Tracking with MIMO sonar systems: applications to harbour surveillance

Abstract

MIMO (Multiple-Input Multiple-Output) sonar systems offer new perspectives for area surveillance especially in complex environments where strong multi-path and dense clutter can become very challenging. This paper proposes a MIMO sonar system based scheme to tackle the difficult problem of harbour surveillance. An emphasis is put on recognition and tracking on low profile mid-water targets. A MIMO simulator which can compute synthetic raw data for any transmitter/receiver pair in multipath and cluttered environment is presented. Moving targets such as boats or AUVs (Autonomous Underwater Vehicle) can also be introduced into the environment. The present paper proposes two radically different methods for the underwater target tracking problem in complex environment: a digital tracker and an analog tracker. On the digital side, an implementation of the recently developed HISP filter (Hypothesised filter for Independent Stochastic Populations) is presented. This filter enables robust multi-object tracking as well as track classification capabilities without the use of heuristics. While digital filters track the objects after processing in the digital domain, an analog filter based on acoustical time reversal techniques enable to track an underwater target using MIMO architecture capabilities is also presented. The proposed modified time reversal technique matches the prediction / data update steps of a traditional tracking filter.

Identification of MultiObject Dynamical Systems: Consistency and Fisher Information

“…One of the advantages with the modified observation model (15) is that the Fisher identity can be utilised as an alternative way of computing the score function based on the unobserved random variables in this model: (16) ∇ θ logp θ (y 1:n ) =Ē θ ∇ θ logp θ (Y α,β 1:n , ς 1:n , D 1:n , X 0:n ) | Y α,β 1:n = y 1:n , wherē p θ (y 1:n , σ 1:n , d 1:n ,…”

“…Making use of the Fisher identity (16), the Fisher information matrix I ∞,0 (θ * ) can be expressed as…”

Improved high‐precision approach based on deconvolution in MIMO sonar imagery

“…By applying waveform diversity in a MIMO array, the MIMO sonar system is capable of acquiring far more degrees of freedom and observation channels than the actual number of transmitting and receiving elements. Inspired by the recent advances of the MIMO imaging radar, the MIMO technique has spread to imaging sonar [3][4][5][6][7].…”

Distributed broadband phased‐MIMO sonar for detection of small targets in shallow water environments

“…MIMO sonar may be viewed as a type of multistatic sonar due to exploiting the spatial diversity gain. Although a general MIMO processing framework is proposed in 2006 [18], MIMO sonar develops slowly due to time‐delay spread, Doppler spread and the limited bandwidth of underwater acoustic channels [19–25]. The orthogonal waveform design has been investigated in [20, 21].…”

“…The space–time waveform diversity is combined with cognition processing to improve the detection performance of the MIMO sonar [19]. Time reversal focusing is utilised for improving the signal‐to‐noise ratio (SNR) required by MIMO detection [22, 23, 25]. The water tank experimental results have validated that the MIMO sonar can generate a high‐resolution image using frequency diversity [24].…”