Firefly algorithm assisted multi-user detection for OFDM/SDMA uplink system

Abstract

Orthogonal frequency division multiplexing/space division multiple access (OFDM/SDMA), as a technique proposed for future high data-rate requirements, is capable of achieving higher user capacity. The error free detection of the transmitted data of the users poses many challenging issues. Alternatives are explored for optimum maximum likelihood (ML) aided detection due to its exponential complexity with the number of users and constellation size. Conventional detection techniques using zero forcing (ZF) and minimum mean square error (MMSE) are less complex with much poor performance. The possibility of nature-inspired meta-heuristic algorithms to detect the user data have been of great research interest these days. With these optimisation algorithms, we can obtain almost optimum results with much reduced computational complexity. In this paper, recently developed firefly algorithm (FA) has been utilised to detect the users' transmitted data and the result has been compared with genetic algorithm (GA) and particle swarm optimisation (PSO) aided data detection.

A novel iterative detection method based on a lattice reduction-aided algorithm for MIMO OFDM systems

“…The system configuration is composed of a receiving user, system bandwidth with 10MHz, the normal CP frame structure, the transmission time interval (TTI) per frame containing 14 OFDM symbols, and the bandwidth configured to 512 subcarriers, where 300 of which are used for data symbol transmission. The other carriers are set to dummy subcarriers with a frequency spacing of 15kHz, modulation types of QAM, receiving and transmitting antenna configurations of 4×4 or 6×6, turbo code using the generated polynomials (7,5) or (11,13), coding rate of 1/2, and log-map decoding mode. for 44  and 66  conditions.…”

“…However, low-complexity detection algorithms are required. The detection algorithms were mainly composed of easy and simple-to-implement methodologies, represented by the zero forcing (ZF) [4] and minimum mean squared error (MMSE) [5], and nonlinear detection algorithms [6]. While each of the ZF and MMSE algorithms could perform signal detection with a low-cost implementation, neither of the detection performance is substantial [7].…”