Ringing In

My third journal paper appeared in the IEEE Transactions on Vehicular Technology last February. You can find a pre-print on arXiv here.

Here are my thoughts on this paper’s strengths and weaknesses.

Strengths:In this paper, we used a simple theoretical model to obtain some practical insights. Specifically, we considered the relay selection problem given that all of the candidate relays employ a partial decode-and-forward strategy. Assuming that the network is static, we showed that the ergodic rate can be maximized by placing all of the candidate relays at a single point. This is intuitively satisfying, as intelligent node placement is essential when initially setting up a wireless network. Also, we were able to use neat mathematical tools such as diversity gain and generalized diversity gain to characterize the performance of relay selection under this partial decode-and-forward framework.

Weaknesses:Unfortunately, neat theoretical models cannot be used to explain all of the practical scenarios that can be realized. In particular, our model only considered the performance impact of path loss, and so lognormal shadowing was not taken into account. Placing a candidate relay node at the ergodic rate-maximizing point appears to be a good strategy, but what if a large building blocks the line-of-sight path between the candidate relay and the intended destination? Also, as far as I can tell, partial decode-and-forward is not employed in either the IEEE 802.16j or the 3GPP LTE-Advanced standards, which limits the practical scope of our work.

This was my second accepted IEEE journal paper, which was another important milestone in my graduate school career. I should also note that the review process for Trans VT is excellent; in particular, this paper was initially submitted in November 2007, and the final acceptance notice was issued in June 2008. This implies that authors seeking a fast turnaround time for their papers should consider submitting to Trans VT. Much of the credit for the smoothness of the review process should go to Weihua Zhuang, who is the Editor-in-Chief of the Transactions.

My second journal paper appeared in the IEEE Transactions on Vehicular Technology last March. You can find a pre-print on arXiv here.

Here are my thoughts on this paper’s strengths and weaknesses.

Strengths:In this paper, we studied a practical problem, namely that of relay selection in a wireless network, and obtained a near-practical solution to this problem. It should be stressed that rate-compatible punctured coding is an important element of industry standards such as IEEE 802.16e and 3GPP LTE. Also, a well-crafted comment from one of the paper’s anonymous reviewers motivated us to thoroughly revise Section 5.1; the results of the revision strengthened our belief in our solution’s near-practicality. In particular, we cited various parameters from the IEEE 802.11a standard to support our assertion that opportunistic relay selection would not significantly increase the level of signaling overhead in a wireless network.

Weaknesses:An inherent limitation to studying a practical problem is the resultant difficulty in obtaining meaningful, closed-form mathematical results. Mathematical elegance is a quality that is rarely attributed to hybrid-ARQ and convolutional coding; combining those two transmission strategies with opportunistic relay selection limited us to the relatively simple analytical results in Section 4. Also, I used an insufficient number of Monte Carlo trials to obtain the simulation results in Sections 5.2 and 5.3. This resulted in plots that would have greatly benefited from either the addition of error bars or additional test runs.

This was my first accepted IEEE journal paper, which was a nice milestone in my graduate school career. I also greatly enjoyed the paper revision process, as it was my first opportunity to mull over reviewer comments and determine how to address them. Formulating a coherent plan for a paper revision is often difficult, but it is well worth the time and effort; for example, I always enjoy reading well-written author responses to my comments whenever I serve as an anonymous reviewer.

My first journal paper appeared in the Journal of Communications and Networks last June. You can find a pre-print on arXiv here.

I thought that I would list some of this paper’s strengths and weaknesses, which may provide some guidance for researchers in the area of wireless communications.

Strengths:We tackled an interesting open problem, which entailed determining the capacity of the full-duplex MIMO relay channel. The end result was a nice technical contribution, where we proposed two partial source-relay cooperation strategies that improved upon a previously proposed lower bound. An inherent advantage in performing information-theoretic research is that reviewers have to be more clever when attempting to reject your submissions, assuming that the stated results are correct. Also, Section 5 illustrates how intuition for technical contributions can be obtained via a few well-thought-out simulation studies.

Weaknesses:An inherent drawback of performing information-theoretic research is that a relatively high bar must be cleared in order to actually obtain the desired results. This issue manifested itself when 1) I made a technical error in the conference version of this paper and 2) I glossed over several critical details in the proofs of our results, which put a crimp in the review process. In retrospect, I should have obtained a more solid grasp of multiuser information theory before tackling this open problem. We were also unable to obtain a sharper outer bound, which limited our options for journal submission; for example, our paper would have been rejected by the IEEE Transactions on Information Theory.

This paper, in some sense, signaled my entry into the world of academia. On a side note, I would like to commend the JCN publication staff, as they did a superb job throughout the paper review and editing process. For interested researchers, the JCN does have the occasional compelling special issue; their guest editors also tend to be fairly well-known in their respective sub-fields.