Boosting power efficiency of physical-layer secured MIMO communications

Lead Research Organisation: Heriot-Watt University
Department Name: Sch of Engineering and Physical Science

Abstract

Today, hardly a week passes by without major incidents of cybercrime, which are constantly encroaching on the security and privacy confidence of each connected individual and the Nation as a whole. The increasingly prevalent wireless information exchanges face even greater challenges as the information is broadcast in open medium and the portable communication devices, e.g. mobile phones and laptops, are unlikely to be equipped with costly and power-hungry cryptographic solutions, especially in the coming quantum era. The physical-layer wireless security has long been regarded as a promising complementary or alternative as it requires little computation capability while endowing systems quantum-immune security. However, to date this is achieved by radiating a significantly more amount of energy in the form of orthogonal artificial noise (AN). This power penalty can be huge, e.g. in some typical application scenarios more than twice as much as the energy is radiated. This, unfortunately, is contrary to the global urgent needs of cutting energy consumption of wireless communication systems and their associated carbon footprint, rendering the current physical-layer wireless security solutions impractical.

This project will be the first systematic study of the physical-layer wireless security under the energy awareness context. We propose to recover the energy penalty of the physical-layer security solutions without compromising security performance. This ambitious vision becomes achievable when a co-design approach, involving transmitter architecture, digital baseband, RF frontend and signal waveforms, is employed. This will require major innovations that currently lie beyond state-of-the-art, which include (a) system architecture- and modulation-aware AN synthesis; (b) non-linear power amplifier-friendly AN synthesis; (c) digital/analogue hybrid modulation and precoding security scheme; and (d) system-level demonstration of physical-layer security solutions. We are not aware of any other research programme that has systematically studied hardware-aware optimum signal waveform synthesis for energy efficient physical-layer wireless security systems.

The partnerships with Toshiba Research Europe (TRE), Ampleon, National Instruments (NI) and Winspread have been specifically established to ensure the successfully delivery of the research programme in every stage. In particular, (a) TRE will provide wireless threat assessment in various application scenarios to ensure the planned research is aligned with the society needs; (b) TRE and Ampleon will provide power amplifier samples and expertise on accurate non-linear power amplifier modelling and characterisation; (c) NI will guide the system integration and demonstration using its USRP platform; (d) Winspread will facilitate small-scale trial of the developed technology through its commercial 4G LTE base-stations.

This two-year research programme will be highlighted through two high-impact practical demonstrators. We firstly intend to show a real-time wireless high-definition video secure wireless transmission in laboratory multipath environment using a bespoke designed physical-layer air interface. The power efficiency improvement will be sufficient to recover the power penalty suffered in the current benchmark physical-layer security solutions. In order to further promote the research outcomes onto the global stage, we plan to integrate the develop security technology onto commercial WiFi and 4G-LTE base-stations.

Planned Impact

This project aims to remove the barriers hindering the large-scale deployment of the physical-layer security in diverse wireless communication systems. Thus, it has direct impact on the wireless cyber security, which is indispensable to realise the future Connected Nation. The physical-layer wireless security plays as the complementary or alternative solution to the current higher-layer cryptographic protocols, and, importantly, it enjoys the ultimate resilience to the forthcoming quantum computing. The technology to be developed in this project and, later, to be deployed beyond the project is able to reduce the data overhead used in the current encryption algorithms and remove the needs of security key management infrastructure. These benefits, together with the recovery of the significant power penalty suffered in the previous physical-layer security designs, will facilitate the industrial uptake, and, in the long-term, fundamentally revolute the wireless cyber security landscape.

On the society aspect, the technology to be developed will significantly improve the power efficiency of physical-layer secured wireless links. This aligns well with the urgent needs of reducing the carbon footprints of the current power-hungry wireless infrastructure. On the other hand, the enhancement of the confidence on security and privacy protection will accelerate the use and acceptance of a multitude of applications that rely on sensitive personal information, such as e-banking, e-commerce and e-healthcare, etc.

Project industrial partners will benefit directly by consolidating their leadership positions. The project insight on hardware-aware physical-layer wireless security will provide Toshiba Research Europe opportunities to incorporate its strength in power amplifier characterisation into the flagship SWAN (Secure Wireless Agile Networks, EPSRC EP/T005572/1) project. The impact of power amplifier behaviours on system performance revealed in the research programme will be fed back to the partner Ampleon. This will provide useful guidance on its future product designs, including power transistors and their associated modelling and pre-distortion algorithms. The USRP platform from the National Instruments will be used differently as the appended bespoke RF frontends will be monitored and optimised. This is distinct to its general use as a fast prototyping tool for implementing various wireless baseband algorithms, wherein RF frontends are transparent to developers. The efforts that will be made to implement the lab demonstrators will illustrate the important values of the USRPs on RF and Microwave research. Lastly, as the major customers of Winspread are private and emergency wireless communication providers, the wireless security is of high importance. Thus, the technology to be delivered in the research project has potential to become the unique selling point of its private network products.

The knowledge generated in this cross-disciplinary project will be fed into the curriculums that will be delivered in the Heriot-Watt UK and oversea campuses. In addition, a number of highly relevant public engagement events (see details in 'Pathways to Impact') are built into the programme. All these will ensure the long-term impact of (a) passing the state-of-the-art knowledge and skills beyond the research team; (b) inspiring future research and industrial leaders; (c) allowing the general public to connect with the research team; and (d) promoting STEM in general. Whenever possible, the research data and outcomes will be deposited in public domains (e.g. University data repository and PI's personal website) to facilitate the validation and the continuity of the future research.

Publications

10 25 50