This paper is devoted to the estimation of target motion parameters with a forward scatter radar (FSR). To provide an upper bound for the estimation performance, a closed-form expression of the Cramer–Rao lower bound (CRLB) is provided for the main target signal parameters, namely Doppler rate, baseline crossing instant, and main lobe width parameter of the target amplitude modulation pattern. This is performed by operating on the signal at the output of the standard FSR processing scheme, namely on the crystal video detector signal: the received signal, which is passed through a square-law envelope detector followed by a DC removal filter. The estimate of the kinematic parameters of moving targets that follow a linear trajectory can be obtained by inverting the relationship between these parameters and the previously estimated target signal parameters. Especially, two FSR configurations are investigated: single-node FSR configuration and dual-baseline FSR configuration with a small angular separation between the two baselines. The closed-form expression of the corresponding CRLB is derived by using a linearized version of the same inverse transformation. Finally, a practical estimator is presented, based on a two-dimensional filter bank, and its performance is compared with the derived CRLB. The results show that unbiased and highly accurate estimates of the kinematic parameters are obtained. The feasibility and the effectiveness of this estimation scheme is demonstrated by applying this technique to recorded live data acquired with a dual-baseline FSR configuration exploiting FM signals as a waveform of opportunity.