Cooperation diversity for OFDM with iterative reception and independent CFO per node

In this paper, an optimized iterative destination node reception approach is introduced, suitable for Orthogonal Frequency Division Multiplexing (OFDM) cooperation diversity with packet-based Forward Error Correction (FEC). The scenario studied represents a multiple-input (MI) situation wherein the multiple inputs are all equal (i.e. repetition coding). Each of the cooperation-involved nodes presents independent carrier frequency offset (CFO) to the destination receiver. The optimized iterative destination receiver employs the Expectation-Maximization (EM) algorithm for inner-loop detection, and an outer-loop Soft-Input Soft-Output (SISO) Viterbi algorithm. Thus, it iteratively estimates the doubly selective channels of the source and relay nodes, compensate for the multiple CFO, and decode the message. Due to the multiplicity of CFO present, the resulting detection algorithm becomes more involved. The novel receiver developed to successfully deal with this situation is termed Cooperative Optimized Independent per Node Receiver (CO-ICFO). The system employs M-ary Phase Shift Keying (MPSK) modulation. A minimum number of embedded pilots is used for initial channel estimation of each link through the Least Squares Estimation (LSE) algorithm. Both systems with and without Automatic Repeat Request (ARQ) are considered. High cooperation diversity gains, on the order of 4 dB are achieved, without ARQ, while the gain with ARQ is reduced to about 3 dB. The presented results show that in general, cooperation offers high gains. Further, adding ARQ results in significantly better performance and more system flexibility.