The number of connected devices is increasing worldwide. Not only in contexts like the Smart City, but also in rural areas, to provide advanced features like smart farming or smart logistics. Thus, wireless network technologies to efﬁciently allocate Internet of Things (IoT) and Machine to Machine (M2M) communications are necessary. Traditional cellular networks like Global System for Mobile communications (GSM) are widely used worldwide for IoT environments.
Nevertheless, Low Power Wide Area Networks (LP-WAN) are becoming wide spread as infrastructure for present and future IoT and M2M applications. Based also on a subscription service, the LP-WAN technology SIGFOXTM may compete with cellular networks in the M2M and IoT communications market, for instance in those projects where deploying the whole communications infrastructure is too complex or expensive. For decision makers to decide the most suitable technology for each speciﬁc application, signal coverage is within the key features.
Unfortunately, besides simulated coverage maps, decision-makers do not have real coverage maps for SIGFOXTM, as they can be found for cellular networks. Thereby, we propose Internet of Things Area Coverage Analyzer (ITHACA), a signal analyzer prototype to provide automated signal coverage maps and analytics for LP-WAN. Experiments performed in the Gran Canaria Island, Spain (with both urban and complex topographic rural environments), returned a real SIGFOXTM service availability above 97% and above 11% more coverage with respect to the company-provided simulated maps. We expect that ITHACA may help decision makers to deploy the most suitable technologies for future IoT and M2M projects.
When an application for IoT is planned to be implemented in a Smart City context, the application designer must select the most appropriate communication technology. On one hand, cellular technologies have been providing subscription-based data connectivity since the GSM technology ﬁrstly opened commercially in 1991. In 2001 the 98% of the West European surface was covered by GSM and its use was mainly focused on voice calls, SMS, and low-speed data services with a bit rate of 9.6 kbps. Within the different cellular technologies, IoT and M2M services can be provided from the second generation (2G) on, with speciﬁc features for IoT and M2M communications expected in the future NB-IoT standard, still to be released.
SERVICE AVAILABILITY IN SIGFOXTM AND GSM
Figure 1 shows the coverage maps from SIGFOXTM (Figure 1a) and the mobile service provider with the largest extension covered (Figure 1b) in the Gran Canaria Island. We can observe how both SIGFOXTM and the mobile service provider cover a similar extension. Besides simulated service maps, carriers provide online tools to check whether a given position from a map has service. Nevertheless, these service maps may not be completely accurate since they are simulated from signal transmission models and using the available topographic information.
PROPOSED LP-WAN SIGNAL RECEIVER PROTOTYPE
Figure 2 represents our prototype composed of a hardware part (ITHACA-device), a server to receive the data from all the base stations (ITHACA-server) and a software tool (ITHACA-tool) which will be described in Sections 4.2–4.4, respectively. Although compatible with other LP-WAN technologies, our prototype has been built to work with the SIGFOXTM technology, a highly used proprietary development of LP-WAN. From now on, we will refer to SIGFOXTM and LP-WAN indistinctly.
Figure 4 represents the procedure of the prototype’s development. The Figure 4a shows all the hardware elements which the ITHACA-device is composed of. The assembly procedure of all the hardware elements on the PCB board is represented in Figure 4b. The Figure 4c displays the assembly between the PCB board (with the communication and control modules), a communication board to transfer the data from the PCB board to the Arduino and, the control board. Finally, the Figure 4d represents the ITHACA-device located inside a plastic shield to protect the hardware elements.
METHODOLOGY AND EXPERIMENTS
Measurement points have been distributed in both urban and rural areas, according to the island characteristics. Figure 10 shows the ofﬁcial area delimitation for both urban and rural environments. In addition, the measurement points in both urban and rural environments are marked on the map with white circles.
Figure 12 represents the Service Map tab from the ITHACA-tool displaying the statistics from given measurements along the ﬁeld trial’s journey. Notice that each measurement is represented in the map by a given color. That is because each color represents a LP-WAN availability service depending on the number of base stations which received the message from the ITHACA-device.
Figure 13 shows the transmissions statistics in rural environments. We can observe how the majority of the transmissions were successfully received by more than three base stations (green) with a 73.5% in respect to all the transmission and 77.78% from all the successfully received for at least one station. On the other hand, only a 5.5% of the transmissions were not received by any station.
The area with available service is represented with color. Opposite, the non-color areas are not covered by the SIGFOXTM network. Both ﬁeld trials and simulation are performed considering the maximum transmission power of 14 dBm. Figure 16 presents the overlapping of the simulated coverage map with the actual ﬁeld trials measurements.
CONCLUSIONS AND FUTURE WORK
Within the challenges of the M2M technology, connecting a large number of devices, and achieving long-range distances, are two of the most relevant ones. The connection of devices allows the development of disruptive IoT applications to enhance services in the Smart City context. The traditional cellular networks provide service covering large areas. Speciﬁcally, the second-generation of cellular networks (a.k.a GSM) has proven to work in IoT scenarios.
The more recent LP-WAN technology covers areas of kilometers with low-power transmissions, in exchange for low throughput rates. Within the LP-WAN technologies, UNB modulation (commercially known as SIGFOXTM) reaches more than 30 km in rural areas. We believe that with the yearly increment of devices, an optimal M2M technology is required to efﬁciently allocate the device-to-device communications. Speciﬁcally, both GSM and SIGFOXTM technologies provide a similar theoretical signal, however, infrastructure and user costs are lower for the latter.
With the recent deployment of SIGFOXTM networks, new opportunities for IoT applications designers have appeared. Thus, in order to provide service to customers, IoT service providers simulate their networks to map approximated coverage areas. However, simulations are not always reliable in complex topographical scenarios due to the terrain shape, hence, providing misleading information to Smart City projects decision makers.
Source: University of Padua
Authors: Raulparada | Daniel Cardenes-tacoronte | Carlos Monzo | Andjoanmelia-Segui