Open AccessNear‐perfect reconstruction short length pulses for FBMC systems: re‐optimising OFDP design via semi‐definite programming
Author(s) -
Kobayashi Ricardo,
Mussi Alex,
Abrão Taufik
Publication year - 2019
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
ISSN - 1751-9683
DOI - 10.1049/iet-spr.2018.5177
Subject(s) - filter bank , computer science , filter (signal processing) , interference (communication) , filter design , prototype filter , convex optimization , mathematical optimization , algorithm , regular polygon , telecommunications , mathematics , computer vision , channel (broadcasting) , geometry
Here, the authors develop a prototype filter design for filter bank multi‐carrier (FBMC) systems focusing on short‐length pulses ( K ≤ 3 ) with near‐perfect reconstruction features, favouring operational conditions for FBMC systems. The proposed design is formulated as the optimisation of the weights of a discrete prototype filter that takes into account high symbol reconstruction rates and desirable spectrum features. Despite not being convex, the formulated optimisation problem is relaxed via semi‐definite programming (SDP), enabling solutions that can over‐perform other popular prototype filters. Numerical results show that the proposed design considering overlapping factors K = { 1 , 2 , 3 } can deliver prototype filters with a good signal‐to‐interference ratio (SIR) versus spectrum performance trade‐off. Indeed, for K ≤ 2 , the proposed methodology delivered prototype filters with the highest SIR levels among the available options. Furthermore, the proposed methodology can be extended and combined with other filter optimisation criteria, enabling to comply with different FBMC design requirements.