A novel approach to digital beamforming

Abstract

For many sonar applications, the sensor outputs of a hydrophone array are sampled at a rate significantly higher than that required for waveform reconstruction when digital beamforming is used. The reason for this is that the number of synchronous, or ’’natural,’’ beampointing directions is proportional to the beamformer input rate. This paper presents an implementation of a digital beamformer that achieves the desired synchronous beams while minimizing the sensor channel sampling rate requirement. The technique employs zero padding of sensor data followed by digital interpolation filters to achieve vernier beamformer delays. Interpolation filtering can be done either at the beamformer input or output to minimize processing requirements. The resulting structure realizes a hardware savings since both A/D converter and cable bandwith requirements can be traded off against digital processing complexity to achieve an optimal partitioning.

Imposing pattern nulls on broadband array responses

“…Hence, the minimum sampling rate required for a null at u 0 is f s = 2f U u 0 2f U : (20) This demonstrates the major disadvantages of time domain nulling: oversampling is required to produce a null that is away from broadside or end-re, and the sensor locations must be quantized to multiples of the fundamental spacing. These problems can beovercome by using interpolation beamforming 22 or fractional sample delay FIR lters. 23,27 However, since FIR lters must be quite long to obtain accurate fractional sample delays, these techniques will result in signi cantly longer sensor lters.…”

Increasing beamsteering directions using polyphase decomposition

“…Pridham and Mucci [3] argued that the scheme of Figure 3 is equal to the scheme in Figure 4 for the case that only one beam is formed. This can be seen by interchanging the filter H and the delays p i T 0 s and placing the filter H and the down-sampling in Figure 3 after the summation.…”

Optimization of vertical and horizontal beamforming kernels on the PowerPC G4 processor with AltiVec technology

“…However, this type of system has many disadvantages due to different programming platforms and unequal 3 performance advances in the two separate technologies. Therefore, many manufacturers of high performance general-purpose processors are integrating sets of native signal processing instructions onto their processor cores to offer solutions requiring fewer processors.…”

“…The current implementation that we are going to build upon adds vertical beamforming to the approach in [3] to enable projection of a 3-D underwater image [2]. In addition, the interpolation for the horizontal beamforming kernel is simplified by using a two-point FIR filter for the digital interpolation.…”

“…This is undesirable because it requires additional bandwidth for the overall system. A practical solution employs digital interpolation with Finite Impulse Response (FIR) interpolation filters to achieve a satisfactory time delay resolution [3]. This solution is shown in Fig.…”