All Projects

Background Large Language Models (LLMs) are rapidly becoming the foundation for intelligent assistants, autonomous systems and interactive applications. However, running advanced AI models requires significant computational resources and often introduces latency that can negatively impact user experience. Future applications such as real-time translation, intelligent transport systems, augmented reality assistants and emergency response copilots will require … Read more

Background The future computing infrastructure of cities may not be fixed. Public transport vehicles, autonomous cars and drones are increasingly being viewed as mobile computing platforms capable of providing services wherever demand arises. These mobile edge platforms offer the possibility of bringing computational resources directly to users, supporting applications such as augmented reality, autonomous vehicles, … Read more

Background Many AI systems require significant time, data and computational resources to learn effective behaviours. This creates a major barrier to deploying intelligent infrastructure across different cities, regions and application domains. Transfer Learning offers the possibility of enabling AI systems to reuse knowledge acquired in one environment and rapidly adapt to new situations. Research Challenge … Read more

Background Artificial intelligence is becoming one of the fastest-growing consumers of computational resources worldwide. As AI systems scale, concerns surrounding energy consumption, sustainability and environmental impact are becoming increasingly significant. Future cities will require intelligent infrastructure capable of supporting advanced AI applications while minimising their carbon footprint. Research Challenge As Large Language Models (LLMs), AI … Read more

Background Most edge computing research assumes a fixed topology of static servers. In reality, cities are in constant motion. Buses, trams, and trains carry passengers — and the devices they depend on — through coverage gaps, congestion hotspots, and unpredictable handoff zones. This project introduces Mobile Edge (ME) servers installed on public transport vehicles as … Read more

Background Centralised placement controllers are simple to reason about but brittle in practice: a single point of failure, limited scalability, and high communication overhead as the network grows. Decentralised approaches — where each edge node makes its own decisions — are more resilient, but risk sub-optimal global outcomes when nodes act purely in self-interest. Multi-Agent … Read more

Background Training a reinforcement learning agent from scratch for each new edge environment is expensive — in time, data, and compute. Yet edge networks vary enormously: a placement policy tuned for a dense urban deployment may perform poorly in a suburban network with different server topologies, user mobility patterns, and service mixes. Transfer Reinforcement Learning … Read more

Integer division and remainder are relatively complicated, expensive operations. Most modern processors have at least one integer multiplier unit that has both low latency (typically between 3 and 5 cycles) and high throughput (typically the unit can complete a multiply operation every cycle). In contrast, most modern processors do not have a fast integer division … Read more

Almost all modern processors provide special-purpose instructions for integer multiplication. However, some applications require integer values that are larger than fit in a single machine word, such as 256 bit integers. We can multiply these big integer values using a sequence of smaller multiplies. For example, multiplying two 256-bit integer values might be completed with … Read more

In some applications, such as many cryptography algorithms, multiplication by a known integer constant is a common operation. It’s well known that integer multiplication with a constant can be computed with a sequence of shifts and adds. Most modern processors already have fast integer multiplication units, so in software it is usually faster to use … Read more