The broad adoption of integrated computing systems through the Internet of Things has led to a need for seamless, low-latency interfaces. Deploying these installations in public spaces, such as gaming areas in airports, presents unique challenges. Traditional input modalities are ill-suited for these environments. Cameras compromise user privacy, and wearables are susceptible to theft, damage or loss. Millimetre wave (mmWave) radars are an ideal sensor for such an interface, as they can operate through non-conductive materials and independently of lighting conditions. Most human sensing solutions for mmWave radars use Deep Learning models, which rely on large data sets for training and have a high computational overhead, limiting their feasibility on resource-constrained edge devices. To address these limitations, we present IAmMuse, a lightweight, signal-processing-based interpretation pipeline for human sensing using mmWave radar technology. We filter and enhance the raw point cloud data, using spatio-temporal density information, as well as kinematic context acquired through a few-shot online learning step. IAmMuse uses this pre-processed data to generate a stabilised prediction, classifying the user’s arm position. We implemented a musical system, controlled through these predictions, as an example application for the technology. The user selects musical notes by moving their arms to either a low, middle, or high position, similar to a conductor. To assess the efficacy of this method, we present a comparative analysis with a State-of-the-Art Human Pose Estimation model. This comparison shows that IAmMuse achieves a classification accuracy 50% higher than the State-of-the-Art model, while using less than 1% of the training data. This thesis validates the viability of non-deep-learning-based interpretation algorithms for human sensing with mmWave radars through a fully functional prototype.

Effect Handler Oriented Programming is a promising new programming paradigm, delivering separation of of concerns with regards to side effects in an otherwise functional environment.
This paper discusses the applicability of this new paradigm to static code analysis programs.
Different code analyzers often have many similar, if not identical pieces of code which could be abstracted away.
This abstraction does not come natural to the programming paradigm of Functional Programming but are quite natural within EHOP.
The current programming languages do not yet seem up to the task of rapid generalization of code and elimination of duplicate pieces of code.
However, the concepts present in EHOP will almost certainly be able to eliminate much of this code reduction once the languages have matured further.
The implicit passing of functionality will also allow for clearer code with less unnecessary visual clutter.