The paper describes an iterative detection scheme for MIMO-OFDM systems. The paper is well-structured and can be followed easly, but it should be improved.  

Major isssues

(1) Language should be improved. The paper can be followed, but there are many errors regarding the use of the plural. I would recommend to have it reviewed by a native speaker.   (2) It is not clear the novelty of the approach.  The only state of the art mentioned is in the brief Introduction. A specific State-of-the-Art section should be added in order to emphasize the advantages and disadvantages of the solution proposed respect to the many alternatives that can be found in the literature.   (3) The communications model is simple, which helps to understand easily the solution proposed and allows for observing better the performance of the detection scheme. However, in practice, transceivers include other blocks (e.g., FECs, scramblers or interleavers)  to reduce errors during communications. Thus, a realistic performance evaluation should be performed taking such blocks into considerations.   (4) In addition to (3), a realistic evaluation should be performed in a realistic channel. In the paper it is not clear which is the channel. In L.63 it is said that H is the Rayleigh fading in the frequency domain,  but it is not clear whether this is a mistake or if it is true that the scheme has been designed exclusively for Rayleigh channels (nothing in the rest in the paper indicates so).   (5) The experiments are well presented, but it is not clear why certain parameters have been chosen:   (5.1) Why use a 128-FFT and a CP of 32 subcarriers? It is just because of the small symbol size? Why not use more subcarriers like most modern wireless do (i.e., 512, 1024 or 2048)?   (5.2) In L.118, What is the meaning of "The number of multi-path is assumed to 7"? Does it mean that the number of channel taps is 7? Can you describe such taps and the channel? Note that such a channel might not seem relevant when comparing the implemented methods against each other, but it is relevant when comparing the proposed approach with others presented in the literature.   (5.3) Why is delta equal to 3/4? In fact, delta is defined in the code of CLLL, but it is not clearly indicated how it influences the system.    

Minor issues

(1) L.98: Table 1 is actually not a table…   (2) Regarding the experiments, although plenty of measurements are shown, it would be interesting to observe the performance of the system in the presence of ISI (or ICI), since this is what justifies the use of OFDM.