Showing papers by "Markku Kojo published in 2018"
01 Dec 2018
TL;DR: This paper proposes Fast-Slow RTO (FASOR) mechanism that takes into account special needs in wireless environments while still properly addressing congestion and runs a series of experiments to confirm that FASOR is able to successfully cope with challenging network conditions such as bufferbloat, high level of congestion, and high link-error rates.
Abstract: The Constrained Application Protocol (CoAP) has been designed to be used on constrained devices such as Internet of Things (IoT) devices. The existing congestion control algorithms for CoAP have known shortcomings in addressing congestion and retaining a good level of performance when link errors occur. In this paper, we propose Fast-Slow RTO (FASOR) mechanism that takes into account special needs in wireless environments while still properly addressing congestion. We run a series of experiments to confirm that FASOR is able to successfully cope with challenging network conditions such as bufferbloat, high level of congestion, and high link-error rates unlike the default and CoCoA congestion control that have severe problems with bufferbloated congestion.
13 citations
16 Jul 2018
TL;DR: An extensive evaluation of the existing IETF CoAP Congestion Control proposals finds that they fail to address congestion properly, particularly in the presence of a bufferbloated bottleneck buffer.
Abstract: A huge number of Internet of Things (IoT) devices are expected to be connected to the Internet in the near future. The Constrained Application Protocol (CoAP) has been increasingly deployed for wide-area IoT communication. It is crucial to understand how the specified CoAP congestion control algorithms perform. We seek an answer to this question by performing an extensive evaluation of the existing IETF CoAP Congestion Control proposals. We find that they fail to address congestion properly, particularly in the presence of a bufferbloated bottleneck buffer. We also fix the problem with a few simple modifications and demonstrate their effectiveness.
7 citations
01 Aug 2018
TL;DR: This work runs an extensive set of experiments in different network settings and reveals that even though CoAP over TCP has its known limitations it scales well and performs even better than expected in certain wireless settings that Co AP over UDP algorithms are specifically designed for.
Abstract: The Constrained Application Protocol (CoAP) is specifically designed for constrained IoT devices and is being rapidly deployed for the communication needs of the IoT devices. CoAP has been specified with its own congestion control algorithms because it runs on top of UDP that does not include any congestion control measures. These algorithms aim at taking into account the specific needs of the IoT communication. The need of running CoAP also over TCP has arised recently and is expected to be increasingly deployed alongside with CoAP over UDP. To understand the benefits and shortcomings of both CoAP over TCP and CoAP over UDP, we run an extensive set of experiments in different network settings and compare the performance of CoAP over TCP to the existing congestion control algorithms for CoAP over UDP. Our results reveal that even though CoAP over TCP has its known limitations it scales well and performs even better than expected in certain wireless settings that CoAP over UDP algorithms are specifically designed for, often even outperforming CoAP over UDP.
3 citations