Research Webzine of the KAIST College of Engineering since 2014
Spring 2025 Vol. 24
Control and utilization of sparsity of transmitting users in a massive random access network significantly improve the system sum throughput
Article | Spring 2015
The rapid proliferation of wireless devices and related services heralds a new era of the Internet of things (IoT), which means the advanced connectivity of a massive number of wireless devices. Offering a massive number of wireless connections poses many new technical challenges. Because limited wireless resources have to be shared with communication entities, without central coordination, the scheduling of a massive number of devices for network access has been one of the key challenges. For example, wireless local area network (WLAN) that is based on CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance) allows only one device to communicate at a time, so mobiles contend for network access. However, due to the distributed nature of this technology, there is a collision of transmissions, which causes high latency as the number of devices increases.
The research group of Professor Wan Choi has developed a new random access scheme based on compressive sensing, which enables a receiver to communicate with multiple devices at a time with considerably small amounts of scheduling resources.
The value of the developed technique comes from the control and utilization of the sparsity of transmitting devices in a massive network. Although a random multiple access network inherently has a sparse nature of transmitting devices due to the low user activity, a distributed scheduling is required to effectively control the sparsity and maximize the sum throughput gain from compressed sensing. There is a tradeoff between the probability of successful device identification with compressed sensing and the achievable sum throughput of a multiple access channel (MAC) since the sum throughput increases with the number of transmitting devices but the successful identification probability does not. The developed distributed scheduling scheme effectively balances them to maximize the sum throughput and offer superior throughput performance compared with conventional contention-based random access schemes.
The scheduling scheme for random multiple access was developed in three different cases of channel knowledge – channel state information at transmitter (CSIT), channel state information at receiver (CSIR), and imperfect channel state information at receiver (ImCSIR). The details of research including mathematical analysis will appear in IEEE Transactions on Wireless Communications. The analytical results from an asymptotic sense provide useful insights on utilizing compressed sensing for maximizing the sum throughput and enable us to understand the impact of system parameters related to the scheduling on the system sum throughput.
In environments with many wireless devices, such as IoT networks, centralized user management is not feasible, but the developed random access scheme can find numerous useful applications and replace conventional CSMA protocols.
When and why do graph neural networks become powerful?
Read moreSmart Warnings: LLM-enabled personalized driver assistance
Read moreExtending the lifespan of next-generation lithium metal batteries with water
Read moreProfessor Ki-Uk Kyung’s research team develops soft shape-morphing actuator capable of rapid 3D transformations
Read moreOxynizer: Non-electric oxygen generator for developing countries
Read more