Artificial Intelligence/Machine Learning Front-End Module for Satcom 5g/6g Wireless Access-Backhaul Transceivers

• Country: United Kingdom
• Summary: Artificial Intelligence/Machine Learning Front-End Module for Satcom 5g/6g Wireless Access-Backhaul Transceivers
• GBT Ref No: 102835883
• Deadline: 04 Sep 2024
• Financier: Other Funding Agencies
• Purchaser Ownership: Government
• Tender Value: Refer Document
• Notice Type: Tender
• Document Ref. No.:
• Purchaser's Detail:
  Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details.

  Login to see details
• Description:
• Tenders are invited for Artificial Intelligence/Machine Learning Front-End Module for Satcom 5g/6g Wireless Access-Backhaul Transceivers (Artes 4.0 Spl 5g/6g 6b.119) The objective of the activity is to develop, implement and test a breadboard of a digital transceiver front-end capable of switching between access and backhaul modes of operation for integrated satcom-terrestrial 5G/6G networks. The transceiver core shall be supported by an AI/ML engine implemented for a set of commercially available hardware and software platforms. The activity shall also provide the testbed to test the AI/ML-based transceiver breadboard and to assess its performance in a laboratory environment.Targeted improvements:Enabling protocol agility and radio flexibility to integrates access-backhaul transceivers supporting densified 5G/6G networking architectures not existing today. Up to 50% reduction in the parts count for the digital transceiver front-end module of 5G/6G-satcom integrated access/backhaul nodes.Description:The integration of satellite telecommunications systems in terrestrial 5G/6G networks envisions new network topologies enabled by the concept of satcom Integrated Access and Backhaul(IAB). Satcom IABnodes will support very flexible network deployments, e.g. meshed, chained, mobile and dynamic, by allowing Radio Access Network (RAN) nodes (gNBs) to functionas an ad-hoc radio relay offering backhaul connectivity to their neighbour nodes, in addition to actingas regular access node for user equipment. In that context, the challenges of densified future 5G/6G networks require further advancements of the integrated satcom element. Generic satcom node products will need to be equipped with a dynamic reconfigurable air-interface flexible enough to respond to different system requirements, operational scenarios, and link conditions.By moving radio functionality into software, and by moving the analogue/digital conversion closer to the air-radio interface at the antenna, a varietyof software defined radio schemes may be invoked to implement critical transceiver functions for a dual access backhaul mode of operation. Moreover, viewing the transceiver front-end functionality as a learning and/or search problem in a large and complex space of possible solutions, allows for heuristic engineering approaches rooted in the mechanisms of evolution, natural genetics, artificial intelligence, and machine learning. The anticipated benefit is the implementation of protocol agnostic front-end transceiver modules of reduced complexity, increased flexibility, and ease of manufacturing.The aim of the activity is to develop, implement and test a breadboard of a digital transceiver front-end based on Artificial Intelligence/Machine Learning (AI/ML) engine to execute critical physical layer signal processing tasks for satcom IAB nodes. While the core AI/ML algorithm development shall remain independentof the final HW/SW platforms (such as FPGA, GPU, CPUs or even cloud computing), the core development shall consider the peculiarityof the final platforms. The targeted digital front-end IAB transceiver breadboard shall handle attributeslike interference over a heavily loaded satellite channel in the presence of linear and non-linear impairments, and support handovers in multi-beam multi-layered satellite systems. The activity shall provide means to verify the AI/ML front-end implementation in the target commercially available platform and test its performance via a testbed in a laboratory environment. In comparison to the current state of art, it will create a reference design for transceiver algorithms, techniques and technologies that does not currently exist.Read more
• Documents:

 Tender Notice