In this paper, we are interested in the problem of joint channel estimation and data detection for multi-input multi-output (MIMO) interleaved frequency division multiple access (IFDMA) systems. Although IFDMA is free from the multiple access interference (MAI), it suffers from intersymbol interference (ISI). MIMO-IFDMA system suffers from both ISI and multi-stream interference (MSI). The block iterative generalized decision feedback equalizer (BI-GDFE) is an iterative and effective interference cancellation scheme which could provide near maximum likelihood (ML) performance with very low complexity. However, BI-GDFE needs the channel state information (CSI). In this paper, we utilize the soft estimates of the transmitted symbols provided by the BI-GDFE to estimate the channel via an expectation maximization (EM)-based algorithm. By doing so, a joint channel estimation and data detection receiver is developed. To evaluate the performance of the proposed channel estimation algorithm, we derive the Cramer-Rao lower bound (CRLB). Computer simulations show that the bit error rate (BER) performance of the proposed joint channel estimation and signal detection receiver can reach the performance of the BI-GDFE with perfect CSI.