Events

Electromagnetic Activities are an essential dimension of modern defence.  Being able to exert control on the electromagnetics space in order to degrade, deny, deceive and disrupt is a key area requiring highly technical solutions.  The problems posed are increasingly complex and involve multi-objective optimisation problems spanning both physical (objects, equipment and humans) and cyber (e.g. software, AI, digital twin) domains.  Biological systems are typically evolved to solve problems that span many domains, trading off energy, mobility and sensor performance.  The nature of electromagnetic problems will be outlined for the benefit of those who are unfamiliar with electromagnetics with the purpose of building cross disciplinary research. 

Operating in the EME is an increasingly complex task as it becomes more congested and contested. Gaining access and superiority in the EME will be enabled by ingenuity in techniques and tools, likely moving beyond our conventional engineering approaches. This project addressed the hypothesis that additional inspiration can be found in nature. By utilising biomimicry and drawing inspiration from biology and biophysics in nature, we may unlock new tools, techniques, and strategies. Next-Generation and Generation-After-Next (GAN) capability enhancement can be in the form of emulating how nature conducts particular activities or strategies, realised using conventional technology, or by leveraging biology directly to deliver effects. 

Details to be confirmed

Artificial Intelligence (AI) is currently at the forefront of technological developments in many sectors, like robotics, manufacturing, services, space, and defence. Deep Learning (DL), as the latest incarnation of AI, is used ubiquitously and it is inexorably replacing more conventional signal processing algorithms and approaches. 

However, despite the considerable advances in recent years, DL requires significant hardware acceleration and resources to be effective, as it is rather computationally expensive. In many practical applications, it is simply not possible nor practical to deploy big power-hungry computing architectures. Cloud-based computing is a possible solution to overcome such issues, but this by itself creates other constraints and limitations, such connectivity requirements and slower reaction times due to communication delays. 

In this context, an attractive alternative is given by Neuromorphic (NM) Engineering, seen as the analogue/digital implementation of biological brain-inspired neural networks. NM systems propagate spikes as means of processing data, with the information being encoded in the timing and rate of spikes generated by each neuron of a so-called spiking neural network (SNN). NM implies also a rethinking of the sensing, as data needs to be generated in spiking format, to be effectively and efficiently processed through SNNs. 

On this premise, the key advantages of NM and SNNs are less computational power required, sparser sensing, more efficient and faster processing, and lower power consumption. Indeed, NM has the potential to be game changing in all those applications that present constraints in terms of computational capabilities, power consumption and reaction time. 

The talk will briefly introduce a range of high frequency circuits and antennas, developed in QMUL that can be reconfigured using liquids. We will then see how the natural flow behaviour of liquids can be utilised to create biomimetic antennas that can maintain their performance when subjected to orientation change. Finally, we will consider the future prospective for this kind of approach.