Kabushiki Kaisha Toshiba

Kabushiki Kaislia Toshiba

Alfred Daniel Wiffen

Gavin Tomas Watkins

A radar return channel is estimated at a radar receiver by processing an existing channel estimate along with a radar return signal for an ideal transmission, to produce an updated estimate. Updating can be carried out based on a determination that the existing estimate has become degraded. Embodiments include using a least squares deconvolution to update polynomials describing a transfer function of the channel estimate.

Radar channel estimation

A full duplex dual polar radar transceiver comprising a dual polarisation radar antenna, a transmission path, a horizontal polarisation receive path, and a vertical polarisation receive path, a first cancellation path connected between the transmission path and the vertical polarisation receive path, and a second cancellation path connected between the transmission path and the horizontal polarisation receive path. Each cancellation path is configured to vary a transmission signal provided by the transmission path by varying at least one of a phase shift, a delay, or an amplitude so as to cancel self-interference on each of the vertical and horizontal polarisation receive paths.

Dual polar full duplex radar transceiver

A method of tuning a self-interference cancellation circuit, wherein the self-interference cancellation circuit comprises a filtering means configured to vary a phase and an amplitude of a transmit signal based on a first filter weight. The method comprising transmitting a first transmit signal, receiving a first received signal in response to transmitting the first transmit signal; estimating, based on the first received signal, a first derivate of a cost function with respect to the first filter weight; estimating, based on the first received signal, a second derivative of the cost function with respect to the first filter weight; determining a first step size based on the second derivative; calculating an updated first filter weight based on the first filter weight, the first step size, and the first derivative; and updating the filtering means based on the updated first filter weight.

Growing community of inventors

Bristol, United Kingdom

Alfred Daniel Wiffen