With the OmniSIG Model Hub, DeepSig continues to lead the way in AI-driven innovations, empowering users to harness the full potential of spectrum sensing technology.
Officially launched on August 1, 2024, the Model Hub is DeepSig’s industry-leading platform designed to transform how wireless developers, researchers, equipment vendors and operators access, manage and deploy AI/ML spectrum sensing models.
Centralizing AI/ML Model Management
The OmniSIG Model Hub serves as a central repository for the powerful models that drive the OmniSIG Engine—an artificial intelligence signal processing and inference software formerly known as OmniSIG Sensor. This software, which leverages AI to enhance radio systems, pushes the boundaries of radio frequency sensing and awareness. With the Model Hub, licensed users have a streamlined and user-friendly platform to store, manage and retrieve pre-trained and customized models from a single, secure location.
The Model Hub also hosts DeepSig-certified models that have undergone extensive testing and validation. Users can download and deploy these certified models with ease, streamlining their deployment processes. This reduces the risk of errors and accelerates the integration of advanced AI/ML spectrum sensing capabilities into their systems.
What is OmniSIG?
OmniSIG® is an industry-leading, AI-based signal detection and classification solution from DeepSig, a pioneer in AI-native wireless communications. This software is hardware platform-agnostic, ensuring broad compatibility across various applications in the wireless communications industry. Going beyond the capabilities of existing spectrum monitoring solutions, OmniSIG uses a customizable deep learning approach that leverages convolutional neural networks and specialized RF architectures, to enhance AI feature learning. This results in the system’s ability to detect and classify signals while understanding the spectrum environment for better analysis and decision-making.
OmniSIG software supports a broad range of signal types, including LTE, 5G, WIFI, and more, with the flexibility to add custom signals. Key features include 100x faster signal classification, handling up to 500MHz bandwidth per GPU and compatibility with common software-defined radios.
“By offering a central model repository, we accelerate the ability of users to leverage powerful deep learning-driven AI/ML spectrum sensing with new signals and bands,” said DeepSig CTO Tim O’Shea in a press release.
O’Shea added that DeepSig’s future vision includes extending the Model Hub to incorporate vetted OmniPHY AI models and fostering collaboration for 5G, 6G, and RAN Digital Twin applications. Additionally, there is potential to bring DeepSig’s advanced spectrum sensing to commercial open RAN radio units, enabling enhanced spectrum sharing and activity monitoring.
Types of Models Available
The OmniSIG Model Hub hosts diverse models, catering to a wide range of signal types and applications. Some of the key models available include:
- Commercial Drone Signals: Detect and classify signals from various commercial drones, enhancing security and monitoring capabilities.
- Push-to-Talk Radios: Efficiently identify and manage signals from push-to-talk communication devices, crucial for tactical and emergency response scenarios.
- Communication Devices: Models for various other communication devices, ensuring robust spectrum management and signal classification.
- Specialized Signals: Tailored models for specific use cases, including industrial IoT signals, military communication bands and more.
DeepSig develops and vets these models, ensuring high reliability and performance. Licensed users can leverage these pre-trained models or customize their own to meet specific requirements, accelerating deployment and reducing development time.
Key Features of the OmniSIG Model Hub
The Model Hub is packed with features designed to enhance the user experience and boost productivity:
- Certified Models: Download and deploy tested and validated DeepSig-certified models with confidence.
- Centralized Model Management: Users can search, manage, and share pre-trained and custom models in one secure location.
- Advanced Search: Find models faster using detailed criteria such as signal types, frequency ranges, or training datasets.
- Enhanced Collaboration: Accelerate development by collaborating with the OmniSIG community, fostering collective efforts and reducing training costs.
Empowering the Community
By fostering greater collaboration among users, the OmniSIG Model Hub enhances collective efforts, accelerates development and impact, and allows users to focus on fine-tuning and deploying models to meet specific needs rapidly. This is a game-changer for OmniSIG Engine users, who can now detect and identify various RF emissions across large instantaneous bandwidths of spectrum in milliseconds. This near real-time reporting of anomalies, changes, and threats significantly enhances operational efficiency and security.
A Redesigned Customer Portal
The launch of the OmniSIG Model Hub coincides with the introduction of DeepSig’s redesigned customer portal. This new portal offers a streamlined experience for license management, accessing training materials and exploring relevant documentation. OmniSIG Engine and OmniSIG Studio customers can now benefit from improved reproducibility, simplified debugging and enhanced model performance monitoring, making it easier than ever to gain valuable spectrum insights.
To explore the capabilities of the OmniSIG Model Hub and how it can revolutionize your approach to spectrum sensing, contact info@deepsig.ai.
Additional Resources:
