Ahmed Fakhfakh

Dr. Ahmed Fakhfakh

orcid

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

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

Representative Publications

View All Publications
27 January 2022 in Electronics
Verified
Journal Article
Open access
PDF file
Meriam Ben AmmarMeriam Ben Ammar
Meriam Ben Ammar
Imed Ben DhaouImed Ben Dhaou
Imed Ben Dhaou
DHDhouha El Houssaini
Dhouha El Houssaini
SSSalwa Sahnoun
Salwa Sahnoun
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
Olfa KanounOlfa Kanoun
Olfa Kanoun
Energy limitations remain a key concern in the development of Internet of Medical Things (IoMT) devices since most of them have limited energy sources, mainly from batteries. Therefore, providing a sustainable and autonomous power supply is essential as it allows continuous energy sensing, flexible positioning, less human intervention, and easy maintenance. In the last few years, extensive investigations have been conducted to develop energy-autonomous systems for the IoMT by implementing energy-harvesting (EH) technologies as a feasible and economically practical alternative to batteries. To this end, various EH-solutions have been developed for wearables to enhance power extraction efficiency, such as integrating resonant energy extraction circuits such as SSHI, S-SSHI, and P-SSHI connected to common energy-storage units to maintain a stable output for charge loads. These circuits enable an increase in the harvested power by 174% compared to the SEH circuit. Although IoMT devices are becoming increasingly powerful and more affordable, some tasks, such as machine-learning algorithms, still require intensive computational resources, leading to higher energy consumption. Offloading computing-intensive tasks from resource-limited user devices to resource-rich fog or cloud layers can effectively address these issues and manage energy consumption. Reinforcement learning, in particular, employs the Q-algorithm, which is an efficient technique for hardware implementation, as well as offloading tasks from wearables to edge devices. For example, the lowest reported power consumption using FPGA technology is 37 mW. Furthermore, the communication cost from wearables to fog devices should not offset the energy savings gained from task migration. This paper provides a comprehensive review of joint energy-harvesting technologies and computation-offloading strategies for the IoMT. Moreover, power supply strategies for wearables, energy-storage techniques, and hardware implementation of the task migration were provided.

17 December 2023 in Computers
Verified
Journal Article
Open access
PDF file
AAAbdullah Al-Abadi
Abdullah Al-Abadi
MMMbarka .Belhaj Mohamed
Mbarka .Belhaj Mohamed
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
In recent years, the combination of wireless body sensor networks (WBSNs) and the Internet ofc Medical Things (IoMT) marked a transformative era in healthcare technology. This combination allowed for the smooth communication between medical devices that enabled the real-time monitoring of patient’s vital signs and health parameters. However, the increased connectivity also introduced security challenges, particularly as they related to the presence of attack nodes. This paper proposed a unique solution, an enhanced random forest classifier with a K-means clustering (ERF-KMC) algorithm, in response to these challenges. The proposed ERF-KMC algorithm combined the accuracy of the enhanced random forest classifier for achieving the best execution time (ERF-ABE) with the clustering capabilities of K-means. This model played a dual role. Initially, the security in IoMT networks was enhanced through the detection of attack messages using ERF-ABE, followed by the classification of attack types, specifically distinguishing between man-in-the-middle (MITM) and distributed denial of service (DDoS) using K-means. This approach facilitated the precise categorization of attacks, enabling the ERF-KMC algorithm to employ appropriate methods for blocking these attack messages effectively. Subsequently, this approach contributed to the improvement of network performance metrics that significantly deteriorated during the attack, including the packet loss rate (PLR), end-to-end delay (E2ED), and throughput. This was achieved through the detection of attack nodes and the subsequent prevention of their entry into the IoMT networks, thereby mitigating potential disruptions and enhancing the overall network efficiency. This study conducted simulations using the Python programming language to assess the performance of the ERF-KMC algorithm in the realm of IoMT, specifically focusing on network performance metrics. In comparison with other algorithms, the ERF-KMC algorithm demonstrated superior efficacy, showcasing its heightened capability in terms of optimizing IoMT network performance as compared to other common algorithms in network security, such as AdaBoost, CatBoost, and random forest. The importance of the ERF-KMC algorithm lies in its security for IoMT networks, as it provides a high-security approach for identifying and preventing MITM and DDoS attacks. Furthermore, improving the network performance metrics to ensure transmitted medical data are accurate and efficient is vital for real-time patient monitoring. This study takes the next step towards enhancing the reliability and security of IoMT systems and advancing the future of connected healthcare technologies.

4 June 2024 in Sensors
Verified
Journal Article
Open access
PDF file
HHHiba Hellara
Hiba Hellara
Rim BarioulRim Barioul
Rim Barioul
SSSalwa Sahnoun
Salwa Sahnoun
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
Olfa KanounOlfa Kanoun
Olfa Kanoun
Effective feature extraction and selection are crucial for the accurate classification and prediction of hand gestures based on electromyographic signals. In this paper, we systematically compare six filter and wrapper feature evaluation methods and investigate their respective impacts on the accuracy of gesture recognition. The investigation is based on several benchmark datasets and one real hand gesture dataset, including 15 hand force exercises collected from 14 healthy subjects using eight commercial sEMG sensors. A total of 37 time- and frequency-domain features were extracted from each sEMG channel. The benchmark dataset revealed that the minimum Redundancy Maximum Relevance (mRMR) feature evaluation method had the poorest performance, resulting in a decrease in classification accuracy. However, the RFE method demonstrated the potential to enhance classification accuracy across most of the datasets. It selected a feature subset comprising 65 features, which led to an accuracy of 97.14%. The Mutual Information (MI) method selected 200 features to reach an accuracy of 97.38%. The Feature Importance (FI) method reached a higher accuracy of 97.62% but selected 140 features. Further investigations have shown that selecting 65 and 75 features with the RFE methods led to an identical accuracy of 97.14%. A thorough examination of the selected features revealed the potential for three additional features from three specific sensors to enhance the classification accuracy to 97.38%. These results highlight the significance of employing an appropriate feature selection method to significantly reduce the number of necessary features while maintaining classification accuracy. They also underscore the necessity for further analysis and refinement to achieve optimal solutions.

25 January 2024 in Sensors
Verified
Journal Article
Open access
PDF file
rihab souissirihab souissi
rihab souissi
SSSalwa Sahnoun
Salwa Sahnoun
MBMohamed Khalil Baazaoui
Mohamed Khalil Baazaoui
Robert FrommRobert Fromm
Robert Fromm
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
Faouzi DerbelFaouzi Derbel
Faouzi Derbel
Indoor localization of a mobile target represents a prominent application within wireless sensor network (WSN), showcasing significant values and scientific interest. Interference, obstacles, and energy consumption are critical challenges for indoor applications and battery replacements. A proposed tracking system deals with several factors such as latency, energy consumption, and accuracy presenting an innovative solution for the mobile localization application. In this paper, a novel algorithm introduces a self-localization algorithm for mobile targets using the wake-up media access control (MAC) protocol. The developed tracking application is based on the trilateration technique with received signal strength indication (RSSI) measurements. Simulations are implemented in the objective modular network testbed in C++ (OMNeT++) discrete event simulator using the C++ programming language, and the RSSI values introduced are based on real indoor measurements. In addition, a determination approach for finding the optimal parameters of RSSI is assigned to implement for the simulation parameters. Simulation results show a significant reduction in power consumption and exceptional accuracy, with an average error of 1.91 m in 90% of cases. This method allows the optimization of overall energy consumption, which consumes only 2.69% during the localization of 100 different positions.

23 September 2024 in Forecasting
Verified
Journal Article
Open access
PDF file
MFMajdi Frikha
Majdi Frikha
KTKhaled Taouil
Khaled Taouil
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
Faouzi DerbelFaouzi Derbel
Faouzi Derbel
Power consumption in the home has grown in recent years as a consequence of the use of varied residential applications. On the other hand, many families are beginning to use renewable energy, such as energy production, energy storage devices, and electric vehicles. As a result, estimating household power demand is necessary for energy consumption monitoring and planning. Power consumption forecasting is a challenging time series prediction topic. Furthermore, conventional forecasting approaches make it difficult to anticipate electric power consumption since it comprises irregular trend components, such as regular seasonal fluctuations. To address this issue, algorithms combining stationary wavelet transform (SWT) with deep learning models have been proposed. The denoised series is fitted with various benchmark models, including Long Short-Term Memory (LSTM), Gated Recurrent Units (GRUs), Bidirectional Gated Recurrent Units (Bi-GRUs), Bidirectional Long Short-Term Memory (Bi-LSTM), and Bidirectional Gated Recurrent Units Long Short-Term Memory (Bi-GRU LSTM) models. The performance of the SWT approach is evaluated using power consumption data at three different time intervals (1 min, 15 min, and 1 h). The performance of these models is evaluated using metrics such as Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE). The SWT/GRU model, utilizing the bior2.4 filter at level 1, has emerged as a highly reliable option for precise power consumption forecasting across various time intervals. It is observed that the bior2.4/GRU model has enhanced accuracy by over 60% compared to the deep learning model alone across all accuracy measures. The findings clearly highlight the success of the SWT denoising technique with the bior2.4 filter in improving the power consumption prediction accuracy.

8 October 2022 in Electronics
Verified
Journal Article
Open access
PDF file
IKIlef Ketata
Ilef Ketata
Sarah OuerghemmiSarah Ouerghemmi
Sarah Ouerghemmi
Ahmed FakhfakhAhmed Fakhfakh
Ahmed Fakhfakh
Faouzi DerbelFaouzi Derbel
Faouzi Derbel
The integration of wireless communication, e.g., in real- or quasi-real-time applications, is related to many challenges such as energy consumption, communication range, quality of service, and reliability. The improvement of wireless sensor networks (WSN) performance starts by enhancing the capabilities of each sensor node. To minimize latencies without increasing energy consumption, wake-up receiver (WuRx) nodes have been introduced in recent works since they can be always-on or power-gated with short latencies by a power consumption in the range of some microwatts. Compared to standard receiver technologies, they are usually characterized by drawbacks in terms of sensitivity. To overcome the limitation of the sensitivity of WuRxs, a design of a low noise amplifier (LNA) with several design specifications is required. The challenging task of the LNA design is to provide equitable trade-off performances such as gain, power consumption, the noise figure, stability, linearity, and impedance matching. The design of fast settling LNA for a duty-cycled WuRx front-end operating at a 868 MHz frequency band is investigated in this work. The paper details the trade-offs between design challenges and illustrates practical considerations for the simulation and implementation of a radio frequency (RF) circuit. The implemented LNA competes with many commercialized designs where it reaches single-stage 12 dB gain at a 1.8 V voltage supply and consumes only a 1.6 mA current. The obtained results could be made tunable by working with off-the-shelf components for different wake-up based application exigencies.