The goal of this project is advancing the rapid growth and deployment of massive multiple input, multiple output (mmimo) open radio units (o-RUS) into current and future networks

This federal Project Grant award, valued at $9,964,628.46, was provided by the National Telecommunications and Information Administration (NTIA) under the Public Wireless Supply Chain Innovation Fund Grant Program (CFDA 11.038). The goal of this 3-year project is to advance the rapid growth and deployment of Massive Multiple Input, Multiple Output (MMIMO) Open Radio Units (O-RUs) into current and future networks.

Key activities include studying trade-offs to implement network beamforming while maintaining O-RU performance, determining requirements for demonstrating network beamforming, and porting the DeepSIG OmniPhy neural receiver to operate on reference hardware from Qualcomm and Fujitsu. The expected outcome is a prototype Manufacturing Readiness Level 4, which will enable O-RU and chipset vendors to implement advanced AI/machine learning functions and foster a healthy, cost-competitive MMIMO O-RU ecosystem. The award aims to make these AI/ML enhancements available industry-wide through standardization efforts with the O-RAN Alliance and 3GPP.

Activities

Multiple trades will be studied to determine how network beamforming (NBF) will be implemented while keeping the o-RU performance within specifications. Requirements for the demonstration of NBF will be determined, specifically, details on channel estimation, beamforming weight calculation and equalization as well as fronthaul requirements will be studied. Additionally, the deepsig omniphy neural receiver will be ported to operate on reference hardware provided by qualcomm and fujitsu.

Expected outcomes

The result of this program will be a prototype manufacturing readiness level (MRL) 4) with the manufacturing risk that hardware changes could be required to process all mmimo layers. Deepsig, fujitsu, and qualcomm are all commercially focused and have the resources to rapidly transition research to commercial o-RUS. The project enables o-RU and chipset vendors to implement advanced artificial intelligence (AI) functions to engender a healthy and cost competitive mmimo o-RU ecosystem. Additionally, deepsig expects this work to make AI enhancements available industrywide.

Intended beneficiaries

Deepsig will collaborate with project partners and others in oran and 3GPP standardization efforts to make AI/machine learning (ML) enhancements available industrywide. Standardization will enable more vendors to operate in the telecom space and support small business efforts.

Subrecipient activities

Deepsig does not list any subrecipients in their application.