Information and Communication Technologies (ICTs) continue to overcome many of the challenges related to wireless sensor monitoring, such as for example the design of smarter embedded processors, the improvement of the network architectures, the development of efficient communication protocols or the maximization of the life cycle autonomy. This work tries to improve the communication link of the data transmission in wireless sensor monitoring. The upstream communication link is usually based on standard IP technologies, but the downstream side is always masked with the proprietary protocols used for the wireless link (like ZigBee, Bluetooth, RFID, etc.). This work presents a novel solution (WebTag) for a direct IP based access to a sensor tag over the Near Field Communication (NFC) technology for secure applications. WebTag allows a direct web access to the sensor tag by means of a standard web browser, it reads the sensor data, configures the sampling rate and implements IP based security policies. It is, definitely, a new step towards the evolution of the Internet of Things paradigm.

The Internet protocol suite is increasingly used on devices with constrained resources that operate as both clients and servers within the Internet of Things paradigm.However, these devices usually apply few-if any-security measures. Therefore, they are vulnerable to network attacks, particularly to denial of service attacks. The well-known SYN flood attack works by filling up the connection queue with fake SYN requests. When the queue is full, new connections cannot be opened until some entries are removed after a time-out. Class 2 constrained devices-according to the RFC 7228-are highly vulnerable to this attack because of their limited available memory, even in low-rate attacks. This paper analyses and compares in a class 2 constrained device the performance of 2 commonly used defence mechanisms (ie,recycle half-open connections and SYN cookies) during a low-rate SYN flood. We first review 2 SYN cookies implementations (ie, Linux and FreeBSD) and compare them with a hybrid approach in a class 2 device. Finally, experimental results prove that the proposed SYN cookies implementation is more effective than recycling the oldest half-open connections. KEYWORDSconstrained devices, denial of service, Internet of Things, network throughput, SYN cookies • Class 0 devices are very constrained sensor-like devices with data/code sizes below 10/100 kB. Commonly, these constrained devices require the help of a gateway to communicate with the Internet. An example of a class 0 device is WISP, 5 a programmable passive RFID tag with integrated sensors. • Class 1 devices are slightly less constrained than class 0; they have about 10/100 kB of data/code sizes. They do not require the help of another network device to communicate with the Internet. They use minimal IP stacks and usually support protocols specifically designed for constrained devices. Most wireless sensor nodes fall into this category (eg, TelosB mote).

Piezoelectric energy harvesting is a promising technology that increases the autonomy of low power IoT devices in scenarios that are subjected to mechanical vibrations. This work shows the potential of this technology to power IoT devices with the energy that is harvested from vibrations occurred during air and road transportation. Adjusting the natural resonance frequency of the piezoelectric generator (PEG) to the mechanical acceleration frequency that has the highest power spectral density is key to increase the harvested energy. Therefore, in this work a commercial PEG is tuned to the best spectrogram frequency of a real vibration signal following a two-phase tuning process. The harvested power generated by the PEG has been validated in real scenarios, providing 2.4 μ Wh during flight (take-off, cruise flight, and landing), 11.3 μ Wh during truck transportation in urban areas, and 4.8 μ Wh during intercity transportation. The PEG has been embedded in an ultra-low power IoT device to validate how much this harvested energy can increase the autonomy in a real scenario that is subjected to similar vibrations. An NFC temperature data logger is developed for perishable products that are transported by air and road transports. The energy harvested by the PEG tuned with the methodology proposed in this work has increased the autonomy of the data logger 16.7% during a real use case of 30 h, which validates the potential of the piezoelectric energy harvesting technology to increase the autonomy of future low power IoT devices used in scenarios with aperiodic vibrations.

No abstract

Mobile Wireless Sensor Networks (MWSN) are usually constrained in energy supply, which makes energy efficiency a key factor to extend the network lifetime. The management of the network topology has been widely used as a mechanism to enhance the lifetime of wireless sensor networks (WSN), and this work presents an alternative to this. Software Defined Networking (SDN) is a well-known technology in data center applications that separates the data and control planes during the network management. This paper proposes a solution based on SDN that optimizes the energy use in MWSN. The network intelligence is placed in a controller that can be accessed through different controller gateways within a MWSN. This network intelligence runs a Topology Control (TC) mechanism to build a backbone of coordinator nodes. Therefore, nodes only need to perform forwarding tasks, they reduce message retransmissions and CPU usage. This results in an improvement of the network lifetime. The performance of the proposed solution is evaluated and compared with a distributed approach using the OMNeT++ simulation framework. Results show that the network lifetime increases when 2 or more controller gateways are used.