arXiv subjects

Kamal Singh

Publications and source records attributed to Kamal Singh.

15 recordsLinked to original sources

arXiv2026

CBL: Compact Encoding of JSON-LD Data using CBOR and Bitmaps for Web of Things

The concept of Web of Things (WoT) merges web technologies with knowledge graphs in the context of Internet of Things. Given its widespread adoption in representing and exchanging structured data online, JSON-LD could be an effective format for WoT. Nevertheless, its verbose nature may present challenges for resource-constrained IoT devices with limited bandwidth and memory capacities. In this paper, we present a novel approach to compactly represent JSON-LD data using the Concise Binary Object Representation (CBOR) and bitmaps. Our proposed method is named as CBL which stands for CBOR, Bitmap and List of Key-value pairs. CBL leverages the ideas from CBOR and HDT to achieve an efficient encoding of JSON-LD data. Results demonstrate that our approach provides savings up to 95.1% in terms of network overhead. This could be especially beneficial for IoT devices exchanging data over wireless networks. Moreover, our approach is more efficient than the current approach known as CBOR-LD, which is used to compact JSON-LD data.

cs.NI↗

arXiv2026

Maximum Spectral Efficiency of Adaptive Coherent Terrestrial FSO Links

Terrestrial free-space optical (FSO) communication systems, while designed to operate on large unlicensed optical bandwidths, are fundamentally power-constrained due to stringent eye-safety regulations. Moreover, channel fluctuations inherent to terrestrial FSO links further reduce the received optical power. Consequently, the achievable signal-to-noiseratio (SNR) per hertz could become limited -- especially for future terrestrial FSO systems based on coherent communications. This necessitates the development of efficient and adaptive communication strategies at both the optical transmitter and receiver. However, a comprehensive assessment of adaptive coherent terrestrial FSO systems remains largely unexplored in the existing literature. This paper investigates terrestrial FSO communication links employing adaptive coherent transmission and synchronous heterodyne detection-based reception operating in the shot-noise-limited regime. Specifically, we propose a novel exact closed-form expression for the average spectral efficiency limit of coherent FSO communication systems with optimal adaptive signaling over gamma-gamma turbulence channels with pointing errors. More importantly, we provide a detailed assessment of the impact of turbulence and pointing error impairments on the coherent FSO system performance, revealing several novel and counterintuitive insights. Extensive numerical results help elucidate the intricacies of analyzing these terrestrial FSO systems and clarify a few misconceptions alluded to in recent related literature.

cs.IT↗

arXiv2026

From GNNs to Symbolic Surrogates via Kolmogorov-Arnold Networks for Delay Prediction

Accurate prediction of flow delay is essential for optimizing and managing modern communication networks. We investigate three levels of modeling for this task. First, we implement a heterogeneous GNN with attention-based message passing, establishing a strong neural baseline. Second, we propose FlowKANet in which Kolmogorov-Arnold Networks replace standard MLP layers, reducing trainable parameters while maintaining competitive predictive performance. FlowKANet integrates KAMP-Attn (Kolmogorov-Arnold Message Passing with Attention), embedding KAN operators directly into message-passing and attention computation. Finally, we distill the model into symbolic surrogate models using block-wise regression, producing closed-form equations that eliminate trainable weights while preserving graph-structured dependencies. The results show that KAN layers provide a favorable trade-off between efficiency and accuracy and that symbolic surrogates emphasize the potential for lightweight deployment and enhanced transparency.

cs.LG↗

arXiv2025

Maximum Spectral Efficiency With Adaptive MQAM Transmissions Over Terrestrial Coherent FSO Links

Coherent free-space optical (FSO) communication is recognized as a key enabler for ultra-high-capacity fronthaul and backhaul links in next-generation wireless networks. Spectrally efficient $M$-ary quadrature amplitude modulation (MQAM) formats are well-suited for these links. However, theoretical analyses of adaptive MQAM transmissions over terrestrial FSO channels remain limited. In this letter, we first derive the spectral efficiency limit of adaptive unconstrained MQAM over gamma-gamma turbulence with pointing error. We then show that adaptive transmissions using only six square MQAM constellations performs close to the theoretical limit (within $0.10$-$0.12$ bits/s/Hz) across a wide range of signal-to-noise ratios and channel conditions.

cs.IT↗

arXiv2025

TSKAN: Interpretable Machine Learning for QoE modeling over Time Series Data

Quality of Experience (QoE) modeling is crucial for optimizing video streaming services to capture the complex relationships between different features and user experience. We propose a novel approach to QoE modeling in video streaming applications using interpretable Machine Learning (ML) techniques over raw time series data. Unlike traditional black-box approaches, our method combines Kolmogorov-Arnold Networks (KANs) as an interpretable readout on top of compact frequency-domain features, allowing us to capture temporal information while retaining a transparent and explainable model. We evaluate our method on popular datasets and demonstrate its enhanced accuracy in QoE prediction, while offering transparency and interpretability.

cs.LG↗

arXiv2025

Interpretable Reinforcement Learning for Load Balancing using Kolmogorov-Arnold Networks

Reinforcement learning (RL) has been increasingly applied to network control problems, such as load balancing. However, existing RL approaches often suffer from lack of interpretability and difficulty in extracting controller equations. In this paper, we propose the use of Kolmogorov-Arnold Networks (KAN) for interpretable RL in network control. We employ a PPO agent with a 1-layer actor KAN model and an MLP Critic network to learn load balancing policies that maximise throughput utility, minimize loss as well as delay. Our approach allows us to extract controller equations from the learned neural networks, providing insights into the decision-making process. We evaluate our approach using different reward functions demonstrating its effectiveness in improving network performance while providing interpretable policies.

cs.LG↗

arXiv2024

A Comprehensive Guide to Enhancing Antibiotic Discovery Using Machine Learning Derived Bio-computation

Traditional drug discovery is a long, expensive, and complex process. Advances in Artificial Intelligence (AI) and Machine Learning (ML) are beginning to change this narrative. Here, we provide a comprehensive overview of different AI and ML tools that can be used to streamline and accelerate the drug discovery process. By using data sets to train ML algorithms, it is possible to discover drugs or drug-like compounds relatively quickly, and efficiently. Additionally, we address limitations in AI-based drug discovery and development, including the scarcity of high-quality data to train AI models and ethical considerations. The growing impact of AI on the pharmaceutical industry is also highlighted. Finally, we discuss how AI and ML can expedite the discovery of new antibiotics to combat the problem of worldwide antimicrobial resistance (AMR).

cs.AI↗

arXiv2024

On The Capacity of Low-Rank Dyadic Fading Channels in the Low-SNR Regime

We characterize the capacity of a low-rank wireless channel with varying fading severity at low signal-to-noise ratios (SNRs). The channel rank deficiency is achieved by incorporating pinhole condition. The capacity degradation with fading severity at high SNRs is well known: the probability of deep fades increases significantly with higher fading severity resulting in poor performance. Our analysis of the dyadic pinhole channel at low-SNR shows a very counter-intuitive result that - \emph{higher fading severity enables higher capacity at sufficiently low SNR}. The underlying reason is that at low SNRs, ergodic capacity depends crucially on the probability distribution of channel peaks (tail distribution); for the pinhole channel, the tail distribution improves with fading severity. This allows a transmitter operating at low SNR to exploit channel peaks `more efficiently' and hence improves spectral efficiency. We derive a new key result quantifying the above dependence for the double-Nakagami-$m$ fading pinhole channel - the capacity ${C} \propto (m_T m_R)^{-1}$ at low SNR, where $m_T m_R$ is the severity parameters (product) of the fadings involved.

cs.IT↗

arXiv2023

Timing-As-A-Service (TAAS): On the role of mobile service providers in the context of integrated Industrial Internet of Things (IIoT)

Traditionally, the production efficiency of a factory floor is evaluated using non-real time objective functions. These are based on scheduling punctuality criteria such as 'earliness (a measure of finishing operations ahead of schedule)' and 'tardiness (a measure of delay in executing certain operations)'. With process automation becoming more and more inevitable due to the emergence of Industry 4.0 / Industry 5.0, real-time objective functions are also gaining popularity. One such criterion is 'synchronization' or 'timing'. In this article, an IoT enabled real-time synchronization approach is presented in the context of Integrated Industrial Internet of Things (IIoT) from a service provider's perspective. A new business and operational model, termed as "Timing-As-A-Service (TAAS)", is introduced. In this model, the role of a service provider is to "supply" synchronization from the mobile networks to the IIoT domain.

cs.NI↗

arXiv2022

Low SNR Capacity of Keyhole MIMO Channel in Nakagami-m Fading With Full CSI

In this paper, we obtain asymptotic expressions for the ergodic capacity of the keyhole multiple-input multiple-output (MIMO) channel at low signal-to-noise ratio (SNR) in independent and identically distributed Nakagami-$m$ fading conditions with perfect channel state information at the transmitter and receiver. We show that the low-SNR capacity of this keyhole MIMO channel scales proportionally as $\frac{\textrm{SNR}}{4} \log^2 \left(1/{\textrm{SNR}}\right)$. Our main contribution is to identify a surprising result that the low-SNR capacity of the MIMO fading channel increases in the presence of keyhole degenerate condition, which is in direct contrast to the well-known MIMO capacity degradation at high SNR under keyhole conditions. To explain why rank-deficient keyhole fading channel outperforms the full-rank MIMO fading channel at sufficiently low-SNR, we remark that the rank of the MIMO channel matrix has no impact in the low-SNR regime and that the double-faded (or double-scattering) nature of the keyhole MIMO channel creates more opportunistic communications at low-SNR when compared with pure MIMO fading channel which leads to increased capacity. Finally, we also show that a simple one-bit channel information based on-off power control achieves this low-SNR capacity; surprisingly, this power adaptation is robust against both moderate and severe fading for a wide range of low SNR values. These results also hold for the keyhole MIMO Rayleigh channel as a special case.

cs.IT↗

arXiv2021

Capacity and Performance Analysis of RIS-Assisted Communication Over Rician Fading Channels

This paper investigates two performance metrics, namely ergodic capacity and symbol error rate, of mmWave communication system assisted by a reconfigurable intelligent surface (RIS). We assume independent and identically distributed (i.i.d.) Rician fadings between user-RIS-Access Point (AP), with RIS surface consisting of passive reflecting elements. First, we derive a new unified closed-form formula for the average symbol error probability of generalised M-QAM/M-PSK signalling over this mmWave link. We then obtain new closed-form expressions for the ergodic capacity with and without channel state information (CSI) at the AP.

cs.IT↗

arXiv2021

Fast Beam Training for RIS-Assisted Uplink Communication

In this work, we propose a beam training codebook for Reconfigurable Intelligent Surface (RIS) assisted mmWave uplink communication. Beam training procedure is important to establish a reliable link between user node and Access point (AP). A codebook based training procedure reduces the search time to obtain best possible phase shift by RIS controller to align incident beam at RIS in the direction of receiving node. We consider a semi passive RIS to assist RIS controller with a feedback of minimum overhead. It is shown that the procedure detects a mobile node with high probability in a short interval of time. Further we use the same codebook at user node to know the desired direction of communication via RIS.

eess.SP↗

arXiv2020

Comments and Corrections to "Capacity of Multiple-Antenna Systems With Both Receiver and Transmitter Channel State Information"

The correspondence cited in the title [1] derived ergodic capacity of the coherent multiple-input multiple-output (MIMO) channel in independent and identically distributed (IID) Rayleigh fading. While the theoretical results are correct, several plots in the paper are incorrect. In this correspondence, we correct the plots. More importantly, the corrected plots present an interesting and compelling contrast between performances of the coherent MIMO systems with and without channel state information at the transmitter; whereas this view is somewhat limited in [1] because of flaws in the capacity curves.

cs.IT↗

arXiv2019

Optimal Power Control in Decentralized Gaussian Multiple Access Channels

We consider the decentralized power optimization problem for Gaussian fast-fading multiple access channel (MAC) so that the average sum-throughput is maximized. In our MAC setup, each transmitter has access to only its own fading coefficient or channel state information (CSI) while the receiver has full CSI available at all instants. Unlike centralized MAC (full CSIT MAC) where the optimal powers are known explicitly, the analytical solution for optimal decentralized powers does not seem feasible. In this letter, we specialize alternating-maximization (AM) method for numerically computing the optimal powers and ergodic capacity of the decentralized MAC for general fading statistics and average power constraints. For illustration, we apply our AM method to compute the capacity of MAC channels with fading distributions such as Rayleigh, Rician etc.

eess.SP↗

arXiv2018

Towards a Question Answering System over the Semantic Web

Thanks to the development of the Semantic Web, a lot of new structured data has become available on the Web in the form of knowledge bases (KBs). Making this valuable data accessible and usable for end-users is one of the main goals of Question Answering (QA) over KBs. Most current QA systems query one KB, in one language (namely English). The existing approaches are not designed to be easily adaptable to new KBs and languages. We first introduce a new approach for translating natural language questions to SPARQL queries. It is able to query several KBs simultaneously, in different languages, and can easily be ported to other KBs and languages. In our evaluation, the impact of our approach is proven using 5 different well-known and large KBs: Wikidata, DBpedia, MusicBrainz, DBLP and Freebase as well as 5 different languages namely English, German, French, Italian and Spanish. Second, we show how we integrated our approach, to make it easily accessible by the research community and by end-users. To summarize, we provided a conceptional solution for multilingual, KB-agnostic Question Answering over the Semantic Web. The provided first approximation validates this concept.

cs.AI↗