We explore how embedded AI can make future IoT systems smarter, more energy-efficient, and responsive to the physical world. 

The ever-increasing demand for computational requirements for emerging applications such as machine learning on the edge, low-power, sustainable, etc, have surpassed the capabilities of today’s computing technologies and in order to cater those application requirements, alternative or unconventional computing architectures and paradigms are essential. Therefore, our vision in this scope is to exploit biological devices and/or design bioinspired circuit architectures emerging technologies to develop next generation sensing and computing. We will invent, design, prototype and demonstrate the potential of such alternative computing and sensing technologies.  

Active Projects:

• Hardware-Design for Harsh Evironments [Two PhD Projects ending 2027]

  • Light-Weight Machine Learning Algorithms (DT2HDL tool)

  • Power-Management Units

• Embedded AI for Intelligent Vehicular Signal Processing [PhD Project ending 2029]

• Smart Wearables [MSc by Research, ending 2027]

• Battery-Less Embedded Systems for IoT [PhD Project]

Past Projects List:

• Unconventional Electronic substrates for sustainable electronics [2022-2023] in collaboration with Unconvensional Computing Lab, UWE Bristol

• e-Assessments in Analogue Electronics [2018-2023] in collaboration with Indunil Sikurajapathi, Karen Henderson and Rhys Gwynllyw, UWE Bristol

• Contact-less cell activity monitoring systems [2018-2021] in collaboration with IBST, UWE Bristol 

• 2.5D Silicon Interposer for Heterogeneous Integration [2013-2016]

• 3D IC integration [2007-2013]

• CMOS Point-of-Care Systems [2012-2016] 

• On-chip interconnet Modelling and Analysis [2004-2008]

• Analogue Mixed Signal Circuit Design [2000-2001]