Shanker Shreejith

Eashan Wadhwa, Georgios Floros, Shanker Shreejith

Pre-implementation behavioural simulation routinely validates functional correctness, yet it also produces rich switching-activity traces that are typically discarded by FPGA computer-aided design (CAD) flows. Prior simulation-guided and power-aware FPGA optimisations demonstrate the promise of exploiting this metadata, but many rely on fixed thresholds, narrow decision heuristics, or limited design awareness, often incurring substantial area overhead. This paper presents Context-aware Simopt-Power, a simulator-guided optimisation framework that combines activity metadata with lightweight structural features (sequential proximity, logic-depth proxies, and fan-out estimates) to more precisely target high-impact regions of the netlist. We additionally remove empirically tuned constants, replacing them with architecture-aware parameters such as LUT size and mapping constraints, and evaluate trade-offs using power, delay, and a more useful metrics, area-delay product (AD) and power-delay product (PD). Implemented in an open-source Yosys/ABC flow and evaluated on the complex Koios deep-learning accelerator benchmarks, Context-aware Simopt-Power achieves an average 6.8% dynamic-power reduction while limiting LUT overhead to 11.2%, thus enabling a holistic design optimisation.

Abhishek Duttagupta, Jin Zhao, Shanker Shreejith

Federated Learning (FL) is a distributed learning scheme that enables deep learning to be applied to sensitive data streams and applications in a privacy-preserving manner. This paper focuses on the use of FL for analyzing smart energy meter data with the aim to achieve comparable accuracy to state-of-the-art methods for load forecasting while ensuring the privacy of individual meter data. We show that with a lightweight fully connected deep neural network, we are able to achieve forecasting accuracy comparable to existing schemes, both at each meter source and at the aggregator, by utilising the FL framework. The use of lightweight models further reduces the energy and resource consumption caused by complex deep-learning models, making this approach ideally suited for deployment across resource-constrained smart meter systems. With our proposed lightweight model, we are able to achieve an overall average load forecasting RMSE of 0.17, with the model having a negligible energy overhead of 50 mWh when performing training and inference on an Arduino Uno platform.

Shashwat Khandelwal, Anneliese Walsh, Shanker Shreejith

In this paper, we explore low-power custom quantised Multi-Layer Perceptrons (MLPs) as an Intrusion Detection System (IDS) for automotive controller area network (CAN). We utilise the FINN framework from AMD/Xilinx to quantise, train and generate hardware IP of our MLP to detect denial of service (DoS) and fuzzying attacks on CAN network, using ZCU104 (XCZU7EV) FPGA as our target ECU architecture with integrated IDS capabilities. Our approach achieves significant improvements in latency (0.12 ms per-message processing latency) and inference energy consumption (0.25 mJ per inference) while achieving similar classification performance as state-of-the-art approaches in the literature.