• 전자통신동향분석
• /
• 제37권5호
• /
• pp.44-53
• /
• 2022

The Internet of Things (IoT) is growing exponentially, with the number of IoT devices multiplying annually. Accordingly, the paradigm is changing from cloud computing to edge computing and even tiny edge computing because of the low latency and cost reduction. Machine learning is also shifting its role from the cloud to edge or tiny edge according to the paradigm shift. However, the fragmented and resource-constrained features of IoT devices have limited the development of artificial intelligence applications. Edge MLaaS (Machine Learning as-a-Service) has been studied to easily and quickly adopt machine learning to products and overcome the device limitations. This paper briefly summarizes what Edge MLaaS is and what element of research it requires.

• IEIE Transactions on Smart Processing and Computing
• /
• 제1권2호
• /
• pp.106-116
• /
• 2012

The huge success of location-aware applications has called for the rapid development of an alternative positioning system to the global positioning system (GPS) for indoor localization based on existing technologies, such as 802.11 wireless networks. This paper proposes the Wireless MAC Processor Positioning System (WMPS), which is a localization system running on off-the-shelf 802.11 Access Points and based on the time-of-flight ranging of users' standard terminals. This paper proves through extensive experiments that the propagation delays can be measured with the accuracy required by indoor applications despite the different noise components that can affect the result: latencies of the hardware transreceivers, multipath, ACK jitters and timer quantization. Key to this solution is the choice of the Wireless MAC Processor architecture, which enables a straightforward implementation of the ranging subsystem directly inside the commercial cards without affecting the basic DCF channel access algorithm. In addition to the proposed measurement framework, this study developed a simple and effective localization algorithm that can work without requiring any preliminary calibration or device characterization. Finally, the architecture allows the measurement methodology to be adjusted as a function of the network load or propagation environments at the run time, without requiring any firmware update.