Phantom limb pain (PLP) is a condition where pain is perceived as arising from a missing limb. Despite the high prevalence of PLP among individuals with amputation, the underlying mechanisms of this painful condition are poorly understood. The aim of the research presented here is to analyse high-resolution EEG data to differentiate oscillatory patterns in amputees who experience PLP from two control groups, amputees without PLP and individuals without amputation, and generate novel insights into the pathophysiology of PLP. This work expands on a previous investigation on the use of common spatial pattern decomposition, to explore brain activity patterns associated with PLP. We apply a Filter Bank Common Spatial Pattern (FBCSP) approach, which allows a more comprehensive and refined analysis across various EEG frequency. Our preliminary analysis revealed different power spectra profile for the population with PLP vs controls. Differences were observed between groups when examining the spatial filters computed for key physiological frequency bands, such as the alpha band (8–13 Hz) and the beta band (14–29 Hz). These observations align with existing theories on the role of these frequencies in chronic pain states and suggest a divergence in neural processing between individuals with and without PLP. The findings underscore the potential of EEG-based spectral analysis in revealing the intricate neural dynamics underlying PLP. While these results offer promising insights, further investigation is necessary to fully understand their biological and clinical implications. Our study highlights the importance of frequency-specific analysis in uncovering the complex neural mechanisms of PLP.