JSJaspreet Singh

Jaspreet Singh

*This is an algorithmically generated author record based on publicly available information.

Jaspreet Singh has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Towards Energy-Efficient AI: Performance Insights from CNN vs SNN on Neuromorphic Systems

    by Jaspreet Singh, Anshu Mehta

    Over the past few decades, artificial intelligence (AI) has gained momentum, which is mainly because of deep learning models that are implemented on GPUs and TPUs. These accelerators have the Von Neumann bottleneck and high-power consumption as limiting factors which do not facilitate the ability of data to flow between processing and memory units even when widely used. To overcome these limitations, neuromorphic computing, which is inspired by the brain, uses massively parallel architectures and event-driven neurons with spiking to implement a paradigm. The following paper provides a performance-based analysis of neuromorphic computing and its relationship with artificial intelligence. Neuromorphic processor systems currently being studied, including the IBM TrueNorth, Intel Loihi and BrainScaleS, are evaluated and tested on benchmark tasks such as gesture recognition and MNIST classification. Experimental evidence suggests that neuromorphic systems are able to retain competitive accuracy rates at power consumption of up to 30 times lower and latency of 2-3 times lower than deep neural networks on a GPU.


  • Optimized Popularity Based Replica Placement (PBRP) for Cloud Storage: Performance Analysis

    by Jaspreet Singh, Tanureet Kaur, Amanpreet Kaur, Divya Singh

    In the era of rapidly growing cloud computing services, efficient data replication strategies are crucial for enhancing system performance, reliability and resource optimization. This research paper highlights concept of scalability, with a particular focus on Popularity Based Replica Placement, as a technique to strengthen scalability. The study compares the Traditional Popularity-Based Replica Placement (PBRP) algorithm with an Optimized PBRP approach. The proposed algorithm improves scalability in cloud computing by dynamically placing the replicas based on their data access patterns. The proposed approach mainly focuses on file popularity and adaptive threshold to ensure replicas are distributed optimally across multiple datacenters. The performance comparison is done to analyze execution time, storage utilization and data availability. For Simulation results we have used CloudSim to demonstrate that optimized PBRP improves load balancing, reduces latency, and optimizes storage usage. This makes PBRA a promising solution for scalable cloud storage systems. The findings are evidence that optimized PBRA is an effective solution for scalable cloud environments as it offers reliability and operational efficiency.

  • Balancing Security and Energy Efficiency in VANETs: Insights and Decision-Making Optimization Workflow

    by Abhishek Gupta, Jaspreet Singh

    Vehicular Ad Hoc Networks (VANETs) have become an essential part of existing intelligent transportation systems, permitting real-time communication between vehicles and infrastructure. However, the integration of VANETs poses significant challenges, particularly in terms of security and energy efficiency, which are paramount for ensuring reliable and sustainable operations. This paper provides a comprehensive review of current research on security mechanisms and energy-efficient strategies in VANETs. Additionally, this paper proposes an evaluation framework based on literature survey to address two goals of enhancing energy efficiency and security in VANETs. The recommended workflow includes decision-making processes for optimizing data availability, ranks providers based on their energy-and time-efficiency, and makes use of machine learning approaches for resource allocation. An analysis on how breaches of security affect power consumption and data transmission indicates that breaches in security lead to higher energy consumption that results from increasing security operations after the breach. The study lays the foundation for future works on the development of adaptive algorithms that takes into consideration both energy-efficient and secure vehicular communication.


  • Energy Efficient Workflow Management System for Workload Distribution and Scheduling in Cloud Data Centers

    by Anu Priya Sharma, Jaspreet Singh

    Cloud computing is deemed appropriate for enterprises because to its adaptability and the delivery of digital amenities through the Internet. The cloud offers virtually unlimited computing resources on request, devoid of initial expenses or extended agreements, allowing enterprises to fulfil their needs for computing more cost-effectively. Several aspects are impacted by the platform's efficacy, including QoS, overload, the lifespan, energy usage, reliability, and adaptability. Data centers significantly contribute to carbon emissions, hence impacting universal carbon footprints. In consideration of delivering an environmentally conscious approach to society, different aspects that can facilitate the attainment of carbon neutrality and optimal sustainability were examined. Throughout the data centre's lifetime, our research indicated that a sustainable approach to incoming workflow was necessary. This research examined numerous bio-inspired strategies to identify the most effective method for load balancing in geographically distributed cloud data centers. The four unique algorithms-ACO, Improved ACO, Firefly, and Improved Firefly-are assessed concerning their energy usage, the frequency of VM migrations, and number of SLA violations. Enhanced Firefly is distinguished as the superior algorithm in all metrics, exhibiting the lowest energy use, the fewest VM migrations, and least SLA violations. ACO demonstrates its lowest efficiency for energy usage and virtual machine migrations, although shows modest service level agreement performance. The improved ACO offers a thorough enhancement in all areas relative to the original ACO. Firefly, despite its energy efficiency, compromises service quality by elevating SLA violations.


  • Open Access

    OCI-OLSR: An Optimized Control Interval-Optimized Link State Routing-Based Efficient Routing Mechanism for Ad-Hoc Networks

    by Jaspreet Singh, Gurpreet Singh, Deepali Gupta, Ghulam Muhammad, Ali Nauman

    MANET (Mobile Ad hoc Networks) functionality is determined by routing protocols’ ability to adjust to atypical changes in information and communication technologies, topological systems, and connection status. Due to interference, node migration, the growth of several pathways, security, and propagation loss, MANET network configurations are dynamic. The proactive routing protocol enhances the message flow utilized in the neighborhood discovery process by using the multipoint relays (MPR) approach. In order to increase the protocol’s effectiveness and efficiency while maintaining the OLSR protocol’s reliability, the research presented in this paper proposed an improved OCI-OLSR (Optimized Control Interval-Optimized Link State Routing) that focuses on better control interval management, an advanced MPR selection process, reducing neighbor hold time as well as decreasing flooding. The suggested proposed protocol was examined using the NS3 simulator, and it was compared to the standard OLSR version and AODV(Ad-hoc On-Demand Routing) routing protocol. According to the analysis’s findings, the suggested system has a lot of promise in terms of a variety of performance metrics under diverse conditions. Overall, the article makes the case that the OCI-OLSR protocol may enhance the performance of the regular OLSR protocol in wireless ad hoc networks by addressing a number of the protocol’s flaws.