Email: m.moradi@northeastern.edu
December 2024: My CV


I am a Post Doctoral Research Associate at the Institute for the Wireless Internet of Things, where I work on Channel Coding. My research interests lie in Information Theory, Quantum Error Correction, Wireless Communications, and Machine Learning.

Previously, I was a PhD student at Bilkent University, working under the supervision of Prof. Erdal Arıkan. The main focus of my study was on polar-like codes. I also have held several postdoctoral research positions at prestigious institutions, collaborating with leading experts in my field.

Note: I am currently seeking opportunities in the job market for either a faculty position or a research role in an industry R&D lab that aligns with my expertise in coding theory, information theory, quantum error correction, and related areas.

  • We solved a fundamental problem in channel coding in our paper, PAC Codes with Bounded-Complexity Sequential Decoding: Pareto Distribution and Code Design, available on arXiv. In this work, we address a key challenge in channel coding related to the computational complexity of sequential decoding. For convolutional codes, it is known that if the number of transmitted data symbols is less than NR0, where R0 is the channel cutoff rate, the decoding complexity remains constant. However, when the data size exceeds NR0, the required computation grows exponentially. PAC codes, which combine polar and convolutional coding, experience this behavior in a different way. After the first step of channel polarization, half of the channels (N/2) are "bad" channels with a cutoff rate R0-, and the other half are "good" channels with a cutoff rate R0+. In an earlier paper, I proved that if the data size in each half of the rate profile exceeds (N/2)R0- and (N/2)R0+, the decoding complexity grows exponentially. In this current work, we have solved the inverse problem— showing that if the data size is below these thresholds, the decoding complexity remains constant. It is important to note the gap between the cutoff rate R0 and the channel capacity I(W). For convolutional codes operating at rates between these values, decoding complexity is exponential. However, after one step of polarization, we observe that the average cutoff rate is higher than R0 (i.e., (R0- + R0+) / 2 ≥ R0), while the average capacity remains unchanged (i.e., (I(W-) + I(W+)) / 2 = I(W)). This implies that after just one polarization step, it is possible to send data at rates closer to the channel capacity with constant decoding complexity. As we further polarize the channel, we can achieve transmission rates even closer to the channel capacity while maintaining constant decoding complexity. Our numerical results show that for small to medium block lengths, PAC codes offer error-correction performance close to theoretical bounds, with a decoding complexity that remains constant. The next question we aim to explore is how to optimally design PAC codes for list decoding.


    • Note: If you believe there are opportunities for collaboration on a topic, please do not hesitate to reach out to me via email.
    Updates:

    • December 2024: We solved a fundamental problem in channel coding in our paper, PAC Codes with Bounded-Complexity Sequential Decoding: Pareto Distribution and Code Design, available on arXiv.

    • August 2024: Moradi, Mohsen, and Hessam Mahdavifar. "On Fast SC-based Polar Decoders: Metric Polarization and a Pruning Technique." arXiv preprint arXiv:2408.03840 (2024).

    • August 2024: Our paper is published in ISIT 2024. Moradi, Mohsen, and David GM Mitchell. "PAC Code Rate-Profile Design Using Search-Constrained Optimization Algorithms."

    • July 2024: I presented a poster on our recent findings in Center for Ubiquitous Connectivity (CUbiC).

    • January 2024: My paper is published in Finite Fields and Their Applications, titled "Polarization-adjusted convolutional (PAC) codes as a concatenation of inner cyclic and outer polar- and Reed-Muller-like codes." In this study, I show that PAC codes are equivalent to a new class of codes consisting of inner cyclic codes and outer polar- and Reed-Muller-like codes. I use the properties of cyclic codes to establish that PAC codes outperform polar- and Reed-Muller-like codes in terms of minimum distance. Cyclic codes are a well-studied class of codes, and the newly introduced concatenated methods yield the favorable weight distribution of PAC codes. One can study the decoding of these codes or explore new ways to investigate the weight distribution of polar-like codes.

    • May 2023: My paper, "Application of Guessing to Sequential Decoding of Polarization-Adjusted Convolutional (PAC) Codes," is published in IEEE Transactions on Communications (TCOM). This paper introduces a necessary criterion for the rate profile of polar-like codes to enable low-complexity sequential decoding. I am working to show that this is also a sufficient condition. This condition can also be applied to the SCL decoding of polar codes.

    • March 2023: Our paper "A tree pruning technique for decoding complexity reduction of polar codes and PAC codes," is published in IEEE Transactions on Communications (TCOM). In this paper, we introduce an online and in-place algorithm that improves the complexity and latency of the SCL decoding of polar-like codes.

    • August 2022: My paper, "Bit-Flipping for Stack Decoding of Polarization-Adjusted Convolutional (PAC) Codes," has been accepted for presentation at the 10th International Workshop on Signal Design and Its Applications in Communications (IWSDA).

    • July 2022: I defended my PhD! You can find my dissertation online.

    • June 2022: Our paper, "Concatenated Reed-Solomon and polarization-adjusted convolutional (PAC) codes," has been accepted for presentation at IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom).

    • September 2021: My paper, "On sequential decoding metric function of polarization-adjusted convolutional (PAC) codes," is published in IEEE Transactions on Communications (TCOM). This paper introduces an optimal metric function for the sequential decoding of PAC codes.

    • December 2020: Our paper, "Performance and Complexity of Sequential Decoding of PAC Codes," compares the performance of PAC codes with convolutional and 5G polar codes. The results demonstrate a significant advantage for PAC codes over both convolutional and 5G polar codes, particularly in short block-length regimes. Additionally, our numerical results indicate that the complexity of sequential decoding for PAC codes follows a Pareto distribution.

    Technical Reviewer for:
    • IEEE Transactions on Communications
    • IEEE Journal on Selected Areas in Communications
    • IEEE Transactions on Information Theory
    • IEEE Transactions on Wireless Communications
    • IEEE Transactions on Vehicular Technology
    • Electronics Journal
    • IEEE Internet of Things Journal
    • Entropy
    • IEEE Communications Letters
    • IEEE Wireless Communications Letters
    • IEEE International Symposium on Information Theory (ISIT)
    • IEEE International Conference on Communications (ICC)
    • IEEE Wireless Communications and Networking Conference (WCNC)
    • IEEE Global Communications Conference
    • International Symposium on Topics in Coding (ISTC)
    • International Conference on Wireless Communications and Signal Processing
    Flag Counter