Hardware Acceleration of Beamforming in a UWB Imaging Unit for Breast Cancer Detection

Abstract

The Ultrawideband (UWB) imaging technique for breast cancer detection is based on the fact that cancerous cells have different dielectric characteristics than healthy tissues. When a UWB pulse in the microwave range strikes a cancerous region, the reflected signal is more intense than the backscatter originating from the surrounding fat tissue. A UWB imaging system consists of transmitters, receivers, and antennas for the RF part, and of a digital back-end for processing the received signals. In this paper we focus on the imaging unit, which elaborates the acquired data and produces 2D or 3D maps of reflected energies. We show that one of the processing tasks, Beamforming, is the most timing critical and cannot be executed in software by a standard microprocessor in a reasonable time. We thus propose a specialized hardware accelerator for it. We design the accelerator in VHDL and test it in an FPGA-based prototype. We also evaluate its performance when implemented on a CMOS 45 nm ASIC technology. The speed-up with respect to a software implementation is on the order of tens to hundreds, depending on the degree of parallelism permitted by the target technology.

Acceleration of microwave imaging algorithms for breast cancer detection via High-Level Synthesis

“…Filtering also compensates path-length dependent Listing 1: MIST Beamforming computational kernel. A discussion on the determination of the parameters in capital letters can be found in the literature [3], [4]. MUSIC-Inspired Reconstruction.…”

“…Profiling. Previous work on the hardware acceleration of MIST showed that filter weights and alignment delays can be computed offline once the system has been characterized and that the air-skin clutter removal step can be implemented sufficiently fast in software [6], [4]. Consequently, we focused our effort on accelerating the beamforming step (Listing 1).…”

Simulation and design of an UWB imaging system for breast cancer detection

“…The effects of the impairments on the receiver performance are characterized in terms of the overall detection system performance, in other words, imaging quality which is specified by visual inspection of the energy maps reconstructed by imaging algorithm [59] [33] and SMR/SCR. As defined in Chapter 2, the tumor can be detected if and only if SMR and SCR are greater than 10 dB and 0 dB, an energy peak appears at the correct tumor position.…”