TCP is the most reliable transport layer protocol that provides reliable data delivery from source to destination node. TCP works well in wired networks but it is assumed that TCP is less preferred for ad-hoc networks. However, for application in ad-hoc networks, TCP can be modified to improve its performance. Various researchers have proposed improvised variants of TCP by only one or two measures. These one or two measures do not seem to be sufficient for proper analysis of improvised version of TCP. So, in this paper, the performance of different TCP versions is investigated with DSDV and AODV routing Protocols. We analyzed various performance measures such as throughput, delay, packet drop, packet delivery ratio and number of acknowledgements. The simulation results are carried out by varying number of nodes in network simulator tool NS2. It is observed that TCP Newreno achieved higher throughput and packet delivery ratio with both AODV and DSDV routing protocols.Whereas TCP Vegas achieved minimum delay and packet loss with both DSDV and AODV protocol. However TCP sack achieved minimum acknowledgment with both AODV and DSDV routing protocols. In this paper the comparison of all these TCP variants shows that TCP Newreno provides better performance with both AODV and DSDV protocols.

While advancing rapidly, Artificial Intelligence still falls short of human intelligence in several key aspects due to inherent limitations in current AI technologies and our understanding of cognition. Humans have an innate ability to understand context, nuances, and subtle cues in communication, which allows us to comprehend jokes, sarcasm, and metaphors. Machines struggle to interpret such contextual information accurately. Humans possess a vast repository of common-sense knowledge that helps us make logical inferences and predictions about the world. Machines lack this innate understanding and often struggle with making sense of situations that humans find trivial. In this article, we review the prospective Machine Intelligence candidates, a review from Prof. Yann LeCun, and other work that can help close this gap between human and machine intelligence. Specifically, we talk about what's lacking with the current AI techniques such as supervised learning, reinforcement learning, self-supervised learning, etc. Then we show how Hierarchical planning-based approaches can help us close that gap and deep-dive into energy-based, latent-variable methods and Joint embedding predictive architecture methods.

No abstract

A MANET is a collection of nodes connected wirelessly that try to converse with each other with no need for any central control or infrastructure establishment. The model of mobility depicts the poignant nature of every node which is mobile in MANETs that is considered to be realistic. It plays a vital role in measuring the performance of MANETs. Mobility is considered to be the prime motive in simulation, because it is a huge influence over the design and network's performance due to limitation in resources and it lead to packet delivery ratio (PDR), varying velocity node energy (NE). Lots of work has been done to improve the above problems. Therefore, there is a requirement of more improvement in this area to enhance overall performance of mobility models. This paper presents a comparative simulation-based analysis of Gauss Markov, Manhattan, and random waypoint mobility models over TCP Newreno that uses a DSDV and AODV routing protocols. Moreover, experiment results and performance analysis have been performed with PDR and NE of the varying number of mobile nodes.

No abstract