InnoComm Enables a Real-Time Interactive Boxing Experience with Edge AI
Smart fitness is evolving beyond instructional video playback into an interactive training experience capable of sensing movements, analyzing performance, and delivering real-time feedback.
InnoComm supported the development of an intelligent boxing fitness system that integrates sensor trackers, Edge AI computing, video streaming, and a human-machine interface. The system enables users to enjoy a more responsive and immersive boxing training experience at home.
The key challenge was not simply improving processing performance. It involved overcoming the original platform's Bluetooth connection limitations while ensuring that data from multiple sensors could be received reliably, synchronized efficiently, and converted into real-time training feedback.
Overcoming Bluetooth Device Connection Limitations
The boxing training system uses multiple sensor trackers to capture punch count, speed, rhythm, and motion data simultaneously. However, the original platform could connect to only four Bluetooth trackers at a time, which was insufficient to support the complete product requirements.
Through low-level software optimization, Bluetooth architecture integration, and system resource management, InnoComm increased the platform's simultaneous connection capacity from four to up to eight trackers—doubling the number of supported tracking devices.
This enhancement involved more than increasing the number of Bluetooth connections. When multiple trackers transmit data simultaneously, the system must also manage device identification, pairing, data synchronization, packet transmission, and connection stability. Any latency or data loss could directly affect the accuracy of the training results.
Maintaining stable connections and low-latency data transmission across eight trackers within limited system resources therefore became one of the project's core technical achievements and a strong demonstration of InnoComm's system integration capabilities.
Enabling Real-Time Motion Analysis
In addition to supporting multiple connected devices, the smart boxing system must process data as each punch occurs. Sending all sensor data to the cloud for analysis could make the training experience vulnerable to network instability and transmission latency.
To address this challenge, InnoComm adopted the MediaTek Genio 3 series as the core computing platform and deployed motion tracking and data analysis directly on the device. Powered by a quad-core Arm Cortex-A53 CPU and integrated GPU, the platform can simultaneously process sensor data, the user interface, and real-time visual content to deliver a smooth and immersive interactive training experience.
With on-device processing, the system can analyze punch metrics and training posture in real time, reducing the delay between movement and feedback. Users can review their training results immediately without waiting for data to be uploaded to the cloud, resulting in a smoother and more responsive workout experience.
Integrating Video Streaming for an Immersive Training Experience
A complete smart fitness product requires more than accurate motion analysis. It must also deliver a stable and intuitive multimedia experience.
InnoComm integrated Wi-Fi, Bluetooth, and HDMI connectivity, allowing the fitness console to receive data from multiple trackers while streaming boxing classes to a TV. Users can follow the instructor on a larger screen while the system continuously records and analyzes their movements in the background.
Extended workouts and continuous data processing also require careful consideration of computing performance, power consumption, and thermal management. Through hardware-software co-design and system-level optimization, the platform maintains stable multitasking performance while improving power efficiency and battery life.
System Integration Defines the Smart Fitness Experience
This smart boxing project demonstrates that the greatest challenge in product development does not lie in a single processor or sensor. The real challenge is ensuring that multi-device connectivity, video streaming, and application software operate reliably together on one platform.
From overcoming the original Bluetooth limitation and expanding support from four to up to eight trackers, to enabling real-time motion analysis and TV streaming, InnoComm delivered more than a hardware platform. The project also involved low-level software optimization, connectivity integration, and complete system validation.
This end-to-end capability—from hardware design to the final user experience—is how InnoComm helps customers transform smart fitness concepts into reliable, deployable products.