MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087126 A) filed by Dr. A. Kowshika; Mr. Lekshmi. R. Nair; Dr. Neha; Dr. Deepthi Kvbl; Dr. Vishnupriya G S; Dr. Sheik Manzoor; Mr. S. Aravinth; and Mr. K. Srinivasan on July 16, 2026, for Hybrid Ai-Blockchain Architecture For Intelligent Wireless Computing Net-Works.
Inventors include Dr. A. Kowshika; Mr. Lekshmi. R. Nair; Dr. Neha; Dr. Deepthi Kvbl; Dr. Vishnupriya G S; Dr. Sheik Manzoor; Mr. S. Aravinth; and Mr. K. Srinivasan.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: Abstract The invention described herein introduces a novel approach to constructing intelligent wireless computing networks that combine elements of artificial intelligence and blockchain technologies. A distinguishing feature of this approach is the organization of information interaction between blockchain nodes through AI modules and the inverse interaction between AI modules and blockchain nodes. Such information interaction can occur at several levels, including on devices (terminals), radio access points, edge servers, and remote servers, with wireless links connecting these levels. Artificial Intelligence modules at different levels solve various tasks, ranging from estimating the quality of individual channels to optimizing resource allocation, detecting anomalies, and making decisions in the face of uncertainty. In addition, some of these modules implement blockchain technology by storing information in the form of time-stamped blocks using selected nodes of the network for validation. Blocks contain information on measurements, parameters, and AI model updates, identities, reservations, and incentives. The validation of entries in blocks is performed by nodes participating in consensus, and the choice of these nodes can be made using AI methods based on their past behavior, the amount of energy, and computational power. Consensus algorithms are also subject to selection based on traffic load estimates and channel characteristics so that intervals between blocks and the number of nodes involved in validation can be variable, depending on current conditions. Thus, by varying the structure of the blockchain, such as the use of partitions for small groups of nodes, the expenditure of energy and computational resources can be reduced. Artificial Intelligence modules can obtain data from the blockchain via an interface that allows queries to be made and results stored in blocks to be used. Information from AI modules, such as updates to their parameters, can also be transported to the blockchain nodes in encrypted form with validation. Smart contracts stored in the form of software code in the blocks define the rules for allocating resources, evaluating data (channel, computation, training), and rewarding or penalizing their use. In turn, the value of the data can be calculated by the appropriate AI modules, which can be used as coefficients in smart contracts to ensure the desired resource allocation policy. The blocks contain information on the changes in parameters (deltas) of AI models that have been validated and proved useful, thus allowing collective learning about the environment with mutually beneficial decisions, despite the lack of trust between individual nodes. The described technology can be used in a wide range of applications, allocating resources among multiple stakeholders, both in dense networks requiring periodic sleep modes for energy saving and in networks with duty cycle limitations, as well as in networks with multiple operators and infrastructure sharing. Due to the interaction between AI and blockchain components, networks using this technology can provide greater autonomy and trust in resource allocation while reducing energy consumption, computational complexity, and costs. This makes it possible to implement intelligent wireless networks that have proven to be quite effective solutions in the context of the growing need for high-density, multioperator infrastructure.
Disclaimer: Curated by HT Syndication.
