Showing 1–6 of 6 results for author: Ahmadi, M J

AugmenTory: A Fast and Flexible Polygon Augmentation Library

Authors: Tanaz Ghahremani, Mohammad Hoseyni, Mohammad Javad Ahmadi, Pouria Mehrabi, Amirhossein Nikoofard

Abstract: Data augmentation is a key technique for addressing the challenge of limited datasets, which have become a major component in the training procedures of image processing. Techniques such as geometric transformations and color space adjustments have been thoroughly tested for their ability to artificially expand training datasets and generate semi-realistic data for training purposes. Data augmenta… ▽ More Data augmentation is a key technique for addressing the challenge of limited datasets, which have become a major component in the training procedures of image processing. Techniques such as geometric transformations and color space adjustments have been thoroughly tested for their ability to artificially expand training datasets and generate semi-realistic data for training purposes. Data augmentation is the most important key to addressing the challenge of limited datasets, which have become a major component of image processing training procedures. Data augmentation techniques, such as geometric transformations and color space adjustments, are thoroughly tested for their ability to artificially expand training datasets and generate semi-realistic data for training purposes. Polygons play a crucial role in instance segmentation and have seen a surge in use across advanced models, such as YOLOv8. Despite their growing popularity, the lack of specialized libraries hampers the polygon-augmentation process. This paper introduces a novel solution to this challenge, embodied in the newly developed AugmenTory library. Notably, AugmenTory offers reduced computational demands in both time and space compared to existing methods. Additionally, the library includes a postprocessing thresholding feature. The AugmenTory package is publicly available on GitHub, where interested users can access the source code: https://github.com/Smartory/AugmenTory △ Less

Submitted 7 May, 2024; originally announced May 2024.

Integrated Sensing and Communications for Unsourced Random Access: Fundamental Limits

Authors: Mohammad Javad Ahmadi, Rafael F. Schaefer, H. Vincent Poor

Abstract: This work considers the problem of integrated sensing and communication (ISAC) with a massive number of unsourced and uncoordinated users. In the proposed model, known as the unsourced ISAC system (UNISAC), all active communication and sensing users share a short frame to transmit their signals, without requiring scheduling with the base station (BS). Hence, the signal received from each user is a… ▽ More This work considers the problem of integrated sensing and communication (ISAC) with a massive number of unsourced and uncoordinated users. In the proposed model, known as the unsourced ISAC system (UNISAC), all active communication and sensing users share a short frame to transmit their signals, without requiring scheduling with the base station (BS). Hence, the signal received from each user is affected by significant interference from numerous interfering users, making it challenging to extract the transmitted signals. UNISAC aims to decode the transmitted message sequences from communication users while simultaneously detect active sensing users, regardless of the identity of the decoded and detected users. In this paper, we derive an achievable performance limit for UNISAC and demonstrate its superiority over conventional approaches such as ALOHA, time-division multiple access, treating interference as noise, and multiple signal classification. Through numerical simulations, we validate the UNISAC's effectiveness in detecting and decoding a large number of users. △ Less

Submitted 1 May, 2024; v1 submitted 30 April, 2024; originally announced April 2024.

Video-based Surgical Skill Assessment using Tree-based Gaussian Process Classifier

Authors: Arefeh Rezaei, Mohammad Javad Ahmadi, Amir Molaei, Hamid. D. Taghirad

Abstract: This paper aims to present a novel pipeline for automated surgical skill assessment using video data and to showcase the effectiveness of the proposed approach in evaluating surgeon proficiency, its potential for targeted training interventions, and quality assurance in surgical departments. The pipeline incorporates a representation flow convolutional neural network and a novel tree-based Gaussia… ▽ More This paper aims to present a novel pipeline for automated surgical skill assessment using video data and to showcase the effectiveness of the proposed approach in evaluating surgeon proficiency, its potential for targeted training interventions, and quality assurance in surgical departments. The pipeline incorporates a representation flow convolutional neural network and a novel tree-based Gaussian process classifier, which is robust to noise, while being computationally efficient. Additionally, new kernels are introduced to enhance accuracy. The performance of the pipeline is evaluated using the JIGSAWS dataset. Comparative analysis with existing literature reveals significant improvement in accuracy and betterment in computation cost. The proposed pipeline contributes to computational efficiency and accuracy improvement in surgical skill assessment using video data. Results of our study based on comments of our colleague surgeons show that the proposed method has the potential to facilitate skill improvement among surgery fellows and enhance patient safety through targeted training interventions and quality assurance in surgical departments. △ Less

Submitted 21 December, 2023; v1 submitted 15 December, 2023; originally announced December 2023.

Comments: 11 pages, 2 figures, journal

Unsourced Random Access Using Multiple Stages of Orthogonal Pilots: MIMO and Single-Antenna Structures

Authors: Mohammad Javad Ahmadi, Mohammad Kazemi, Tolga M. Duman

Abstract: We study the problem of unsourced random access (URA) over Rayleigh block-fading channels with a receiver equipped with multiple antennas. We propose a slotted structure with multiple stages of orthogonal pilots, each of which is randomly picked from a codebook. In the proposed signaling structure, each user encodes its message using a polar code and appends it to the selected pilot sequences to c… ▽ More We study the problem of unsourced random access (URA) over Rayleigh block-fading channels with a receiver equipped with multiple antennas. We propose a slotted structure with multiple stages of orthogonal pilots, each of which is randomly picked from a codebook. In the proposed signaling structure, each user encodes its message using a polar code and appends it to the selected pilot sequences to construct its transmitted signal. Accordingly, the transmitted signal is composed of multiple orthogonal pilot parts and a polar-coded part, which is sent through a randomly selected slot. The performance of the proposed scheme is further improved by randomly dividing users into different groups each having a unique interleaver-power pair. We also apply the idea of multiple stages of orthogonal pilots to the case of a single receive antenna. In all the set-ups, we use an iterative approach for decoding the transmitted messages along with a suitable successive interference cancellation technique. The use of orthogonal pilots and the slotted structure lead to improved accuracy and reduced computational complexity in the proposed set-ups, and make the implementation with short blocklengths more viable. Performance of the proposed set-ups is illustrated via extensive simulation results which show that the proposed set-ups with multiple antennas perform better than the existing MIMO URA solutions for both short and large blocklengths, and that the proposed single-antenna set-ups are superior to the existing single-antenna URA schemes. △ Less

Submitted 14 July, 2023; originally announced July 2023.

Unsourced Random Access with a Massive MIMO Receiver Using Multiple Stages of Orthogonal Pilots

Authors: Mohammad Javad Ahmadi, Tolga M. Duman

Abstract: We study the problem of unsourced random access (URA) over Rayleigh block-fading channels with a receiver equipped with multiple antennas. We employ multiple stages of orthogonal pilots, each of which is randomly picked from a codebook. In the proposed scheme, each user encodes its message using a polar code and appends it to the selected pilot sequences to construct its transmitted signal. Accord… ▽ More We study the problem of unsourced random access (URA) over Rayleigh block-fading channels with a receiver equipped with multiple antennas. We employ multiple stages of orthogonal pilots, each of which is randomly picked from a codebook. In the proposed scheme, each user encodes its message using a polar code and appends it to the selected pilot sequences to construct its transmitted signal. Accordingly, the received signal consists of superposition of the users' signals each composed of multiple orthogonal pilot parts and a polar coded part. We use an iterative approach for decoding the transmitted messages along with a suitable successive interference cancellation scheme. Performance of the proposed scheme is illustrated via extensive set of simulation results which show that it significantly outperforms the existing approaches for URA over multiple-input multiple-output fading channels. △ Less

Submitted 19 May, 2022; v1 submitted 3 February, 2022; originally announced February 2022.

Random Spreading for Unsourced MAC with Power Diversity

Authors: Mohammad Javad Ahmadi, Tolga M. Duman

Abstract: We propose an improvement of the random spreading approach with polar codes for unsourced multiple access. Each user encodes its message by a polar code, and the coded bits are then spread using a random spreading sequence. The proposed approach divides the active users into different groups, and employs different power levels for each group in such a way that the average power constraint is satis… ▽ More We propose an improvement of the random spreading approach with polar codes for unsourced multiple access. Each user encodes its message by a polar code, and the coded bits are then spread using a random spreading sequence. The proposed approach divides the active users into different groups, and employs different power levels for each group in such a way that the average power constraint is satisfied. We formulate and solve an optimization problem to determine the number of groups, and the number of users and power level of each group. Extensive simulations show that the proposed approach outperforms the existing methods, especially when the number of active users is large. △ Less

Submitted 27 August, 2021; v1 submitted 25 April, 2021; originally announced April 2021.