A Look Back

Simulated Interference Cancellation and Block Modulation (ICBM) coded and uncoded BER with the precoders designed in A’s paper.

Compared ICBM with reuse1 and found that reuse1 performs better than ICBM (by then end of January).

After this work.

And due to my foot in mouth diesease, I raised a question to Prof. G why not we try 3D constellation, and he asked me to work on Multidimensional constellation, which in the beginning seemed interesting but I could hardly find literature on the Multidimensional constellation which I wanted to learn. So, I lost hope in it and gave up.

From here,

My work got parted in many ways, first I was told to design precoders using ICBM paper for 2Rx antenna (SIMO) as the maximum SpAC for 2Rx antenna is 3/4. I tried designing precoders for SIMO using ICBM paper, but I did not find that work interesting, and also I was apathetic in working on it.

Then I leafed through few papers on Space Time Code blocks, and literature on Spatial Modulation which I found very interesting. But I could not find ways to interlink this Spatial Modulation scheme to ICBM.

Dr. H (Ph.D student) told me to come up with precoder design which is a function of channel (ICBM and B-FIA precoders are not a function of channel) for some other schemes which Prof. G suggested, so after thorough search in literature on precoder design, I found a paper exploiting maxizing channel cut-off rate to design precoders in contrast to many papers which exploit Mutual Information as the basis.

I initiated working on channel cut-off rate (not clearly understood the concept of channel cut-off rate) which employs steepest descent algorithm for designing precoders. I am not very certain about its optimality though.

Due to illness, and followed by Quizzes my work got further delayed.

After couple of weeks, to understand the channel cut-off rate I approached Dr. A T, his anwer was “Cut-off rate is not an elegant way to design precoders but try using something like Outage”. Approaching Dr. A P I got clear understanding of channel cut-off rate, and also superficial understanding of outage probability, and the way it is used in designing precoders.

But none paid heed towards it.

I met Prof. G along with Dr. H almost after two weeks, Prof. G then told me to work on intermediate scheme of FIA and B-FIA, and also SIMO case of FIA. FIA uses complete channel information and it is designed in FIA paper for SISO and MIMO. B-FIA, unlike FIA doesn’t need any channel state information. By name itself it is “blind”. But the SpAC ( Symbols per Antenna per Channel use is less in the case of B-FIA in comparision to FIA).

So, now I gotta design precoders with partial channel state information at the transmitter based on the SpAC value. I suppose SpAC with partial CSIT would lie somewhere between SpAC value with perfect CSIT and no CSIT.


Leave a Comment

Your email address will not be published. Required fields are marked *