One of the most important components on any Radio Frequency (RF) transmitter is the power amplifier (PA). PA’s consume most of the transmitters’ energy; therefore, improving their efficiency is of primary concern. A PA is inherently a non-linear device; therefore, it introduces distortion into the transmitted signal, which must be mitigated in order to provide the high levels of performance and efficiency required in mobile handset applications.
The stringent requirements of 5G mobile communication makes the challenges mentioned above more difficult. In this research area we are investigating the design of tunable and multi-band PA’s for the new, high-data-rate 5G technologies for frequencies below 6GHz. We analyse tunable matching networks to determine their coverage and optimize them for tunable PA’s. We also investigate mathematical optimization approaches for the design of multi-band and tunable PA’s.
Moreover, with the development of ever more advanced radio hardware comes the requirement to measure and characterise these systems. Characterisation is a vital part of the innovation process, and communications hardware evolves toward higher and higher levels of performance, we again require more and more advanced characterisation techniques to keep up.
Researchers in the Communication Systems & Networks (CSN) Research Group are developing advanced novel characterisation techniques with the aim of increasing the effective sampling speed and bandwidth of characterisation systems, thus enabling us to fully and accurately characterise the advanced radio frequency power amplifiers being designed to support higher data rate services in 5G, both at sub-6GHz frequencies and millimetre-wave. The novel measurement techniques developed in the CSN lab offer increased characterisation bandwidth and dynamic range at a fraction of the cost of existing approaches.
The increasing complexity of radio front-end designs has been driven by the evolving requirements of communication system standards. One of the greatest current challenges is the proposal of 5G carrier aggregation schemes for mobile communication which requires the receiver to operate with up to five multiple bands concurrently across 0.4 – 6GHz. This poses a significant challenge for the low noise amplifier (LNA) in terms of operating bandwidth and interference rejection. The current solution of using multiple narrowband LNA’s quickly loses its appeal when parameters such as size, cost and power consumption are taken into account. Research ongoing in CSN aims to address the practical issues involved with the implementation of CA schemes, producing blocker resilient low noise amplifiers for tunable multiband receivers, using both commercially available and experimental integrated circuit fabrication methods.
Antennas, being the interface between radio waves propagating in the environment and the electronic circuits, which transmit and receive these signals, are a fundamental part of any wireless communication system. The performance of the antenna can have a dramatic impact on the performance of the communications system, with reduced antenna efficiency leading to increased transmitter power consumption and reduced receiver sensitivity. However, in mobile devices, the requirement for small antenna volume, and multi-band capability, means that efficiency if often traded to meet the requirements for thinner and sleeker handset form factors. Research in CSN is focussed on tunable and reconfigurable antennas for handset applications, using novel materials and tuning technologies to create efficient, small volume antennas for 4G and 5G devices. The CSN lab is equipped with a far field anechoic chamber, allowing the efficiency and 3D antenna pattern of novel design to be fully characterised.
The duplexer connects the transmitter and receiver to a shared antenna, allowing signals to be transmitted and received through a shared antenna, whilst also isolating the receiver from the transmitter. This isolation is required in order to prevent the transmitted signal swamping the receiver with unwanted self-interference, which, being orders or magnitude more powerful than the desired can “drown out” the desired received signal, preventing it from being decoded. Self-interference cancellation uses knowledge of the transmit signal to estimate and subtract the self-interference. Novel self-interference cancellation techniques being developed in the CSN lab have been shown to cancel self-interference by a factor of over 100 dB by combining passive radio frequency cancellation, active radio frequency cancellation, and non-linear digital baseband cancellation. These techniques can also be implemented using low-cost small form factor technologies suitable for deployment in future mobile devices, and allow devices to use the same frequency at the same time for transmitting and receiving, offering up to double the capacity compared to today’s radio systems, which instead use either time-division duplexing (TDD) or frequency-division duplexing (FDD).