دانلود رایگان مقاله انگلیسی اینترنت اشیای بی سیم- هزینه کافی برای انتقال داده TCP در ارتباط رادیویی - IEEE 2017

Abstract

IoT wireless connectivity often presents a lossy and error prone radio link layer between the wireless edge device and the wireless base station. Data transacted over the radio link layer often uses TCP, which provides programmatic error mitigation elements, e.g., loss of and out-of-order packet detection, combined with ACK/NACK from far end to trigger packet retransmission. In the use case of subscription based wireless connectivity, such as cellular or satellite, the information payload is often metered and billed per amount of data passed over the radio link. Hence, a wireless edge device can incur significant data overage charges to transmit a given amount of application information due to the increased billable TCP-driven retransmissions. The aim of this paper is to quantify the TCP-retransmission rate caused by a lossy radio link as it is experienced by a typical IoT LTE wireless device. The authors also propose the creation of a functional layer between the application layer and the lower layers on the wireless edge device, referred to as “RF Fidelity Layer”. Its purpose is to provide real-time situational knowledge of the radio link layer to the application software, which in turn can “tune” the information flow for maximum efficiency and minimum billable data overage due to TCP-driven retransmissions.

V. CONCLUSION AND FUTURE WORK

This paper presented preliminary lab testing to quantify the TCP-retransmission rate caused by a lossy radio link as it is experienced by a typical IoT LTE wireless device. These results indicate that the proposed RF Fidelity Layer is a promising approach to lowering subscriber overcharges and unnecessary battery consumptions at the wireless device. A complementary RF Fidelity Layer at the host is also proposed, which can discern the fidelity of the radio link based upon analysis of data classifications from the wireless device, hence providing similar benefits when carrying out unbidden host communications. While these results indicate that the proposed research method has merit, the range and sensitivity to radio link fidelity may be improved with additional radio stack probes. The next phase of research is going to expand the scope of available probes into the algorithmic fidelity prediction procedure focusing for example upon developing additional radio stack probe information to incorporate into the predictive algorithm in order to improve the detectable band of degraded wireless link fidelity. Finally, collection of real-world field data ࡳ by building out test units that can log real-time probe information and send out measured TCP messages of various sizes to a back-end host ࡳ will be used to validate the proposed concept in the field. Comparisons between various LTE module manufactures will also be studied. This validation will enable synchronization of the TCP traffic traces collected at the host router using Wire Shark with the carrier billing record. Common RF interference sources will be field tested to quantify the impact on TCP data overages with and without the proposed RF Fidelity Layer. One such source of interference can be a nearby brushed motor as it was experienced during the lab testing when a drill press was turned on in an adjoining room. Even in the isolated screen room the testing was adversely impacted with higher than expected TCP errors and overages.