Dmitry Bankov

Mathematical model of LoRaWAN channel access

While 3GPP has been developing NB-IoT, the market of Low Power Wide Area Networks has been mastered by cheap and simple Sigfox and LoRa/LoRaWAN technologies. Being positioned as having an open standard, LoRaWAN has attracted also much interest from the research community. Specifically, many papers address the efficiency of its PHY layer. However MAC is still underinvestigated. Existing studies of LoRaWAN do not take into account the acknowledgement and retransmission policy, which may lead to incorrect results. In this paper, we carefully take into account the peculiarities of LoRaWAN transmission retries and show that it is the weakest issue of this technology, which significantly increases failure probability for retries. The main contribution of the paper is a mathematical model which accurately estimates how packet error rate depends on the offered load. In contrast to other papers, which evaluate LoRaWAN capacity just as the maximal throughput, our model can be used to find the maximal load, which allows reliable packet delivery.

Beacons in dense Wi-Fi networks: How to befriend with neighbors in the 5G world?

To address 5G challenges, IEEE 802.11 is currently developing new amendments to the Wi-Fi standard, the most promising of which is 802.11ax. A key scenario considered by the developers of this amendment is dense and overlapped networks typically present in residential buildings, offices, airports, stadiums, and other places of a modern city. Being crucial for Wi-Fi hotspots, the hidden station problem becomes even more challenging for dense and overlapped networks, where even access points (APs) can be hidden. In this case, user stations can experience continuous collisions of beacons sent by different APs, which can cause disassociation and break Internet access. In this paper, we show that beacon collisions are rather typical for residential networks and may lead to unexpected and irreproducible malfunction. We investigate how often beacon collisions occur, and describe a number of mechanisms which can be used to avoid beacon collisions in dense deployment. Specifically, we pay much attention to those mechanisms which are currently under consideration of the IEEE 802.11ax group.

The study of the distributed control method to hasten link set-up in IEEE 802.11ah networks

Following the interest of the global research and industrial community in the concept of Internet of Things, IEEE 802 LAN/MAN standard committee is developing the IEEE 802.11ah amendment that will shift the Wi-Fi technology to the area of Machine-to-Machine communications. One of the most important problems considered by the new amendment is the energy-efficient communication of wireless stations, the number of which can reach thousands per Access Point. Serving such a large number of stations is not a problem itself, because it can be managed by efficient scheduling, but the usage of different advanced control methods requires association. Before the association, the Access Point knows neither the capabilities of the stations, nor their identifiers, therefore it has to rely on the basic random channel access for setting up the link. When number of connecting statios is large, the contention for channel access is inevitable, which can drastically increase the link set-up time. To hasten the link set-up, IEEE 802.11ah proposes the Distributed Authentication Control protocol. This pioneering paper describes a mathematical model of the Distributed Authentication Control that can be used to find the link set-up time and the parameters that minimize it.

What Is the Fastest Way to Connect Stations to a Wi-Fi HaLow Network?

Wi-Fi HaLow is an adaptation of the widespread Wi-Fi technology for the Internet of Things scenarios. Such scenarios often involve numerous wireless stations connected to a shared channel, and contention for the channel significantly affects the performance in such networks. Wi-Fi HaLow contains numerous solutions aimed at handling the contention between stations, two of which, namely, the Centralized Authentication Control (CAC) and the Distributed Authentication Control (DAC), address the contention reduction during the link set-up process. The link set-up process is special because the access point knows nothing of the connecting stations and its means of control of these stations are very limited. While DAC is self-adaptive, CAC does require an algorithm to dynamically control its parameters. Being just a framework, the Wi-Fi HaLow standard neither specifies such an algorithm nor recommends which protocol, CAC or DAC, is more suitable in a given situation. In this paper, we solve both issues by developing a novel robust close-to-optimal algorithm for CAC and compare CAC and DAC in a vast set of experiments.

Fast centralized authentication in Wi-Fi HaLow networks

In early 2016 Wi-Fi Alliance announced a new technology — Wi-Fi HaLow — which aims to master the raising Internet of Things market with its tremendous number of devices. This technology brings revolutionary changes to Wi-Fi, improving transmission reliability and power efficiency in scenarios with thousands of sensor stations being connected to a single access point. However neither novel channel access methods, nor advanced power management schemes can be used until the stations are connected to the access point, because the access point may not know that the stations exist, to say nothing about their capabilities. Nevertheless, it may happen that a large group of sensor stations are simultaneously trying to connect to the access point, which inevitably leads to continuous collisions and finally to extremely long connection process. This problem has been addressed by the technology developers which designed the Centralized Authentication Control protocol. The protocol provides the access point with an opportunity to hasten the process by managing the ratio of stations which are allowed to start establishing a connection. In this paper, we study how the number of successes depends on this ratio and develop two algorithms, which hasten connection process close to the theoretical limit.

Improving Efficiency of Heterogeneous Wi-Fi Networks with Energy-Limited Devices

A heterogeneous wireless network consists of various devices that generate different types of traffic with heterogeneous requirements for bandwidth, maximal delay and energy consumption. An example of such networks is a Wi-Fi HaLow network that serves a big number of Machine Type Communication battery-powered devices and several offloading client stations. The first type of devices requires an energy-efficient data transmission protocol, while the second one demands high throughput. In this paper, we consider a mechanism that allocates a special time interval (Protected Interval) inside of which only battery powered-powered devices can transmit. We show that appropriate selection of the Protected Interval duration allows battery-powered devices to consume almost the minimal possible amount of energy on the one hand, and to provide almost the maximal throughput for offloading stations on the other hand. To find such duration, we develop a mathematical model of data transmission in a heterogeneous Wi-Fi network.

Is it worth to predict overflows during video streaming over wireless networks

No. In the paper, we consider compressed video streaming over the wireless channel. From time to time, dynamically changing channel characteristics cause queue overflows, packet drops and, finally, video quality degradation. To eliminate it, the well known I-Frame delay algorithm drops the least important packets when the queue overflows. However, this algorithm starts working only when the queue overflows. Inspired by many works on video bitrate and channel quality prediction, we study how I-Frame Delay would benefit if it knew about future queue overflows and filtered packets in advance. We show that even with the ideal prediction the improvement is negligible.