Recurrence structures in univariate time series are challenging to detect. We propose a combination of recurrence and symbolic analysis in order to identify such structures in a univariate signal. This method allows to obtain symbolic representation of the signal and quantify it by calculating its complexity measure. To this end, we propose a novel method of phase space reconstruction based on the signal's time-frequency representation and show that the proposed method outperforms conventional phase space reconstruction by delay embedding techniques. We evaluate our method on synthetic data and show its application to experimental EEG signals.