Nowadays, the transition of Wireless Sensor Networks (WSNs) to Internet Protocol version 6 (IPv6), in particular to IPv6 over Low power Wireless Personal Area Networks (6LoWPAN), is evident [1]. However, in most commonly used implementations, not all IPv6 features are available. For example, current implementations are not very optimized for multicast, despite the many benefits multicast can offer with respect to the number of radio transmissions and the amount of consumed energy.In this paper we present Bidirectional Multicast RPL Forwarding (BMRF), a new multicast protocol that combines the best features of the Routing Protocol for Low Power and Lossy Networks (RPL) multicast on the one hand and of Stateless Multicast RPL Forwarding (SMRF) on the other hand. The main features are bidirectionality and the ability to offer a choice between Link Layer broadcast and Link Layer unicast for which the threshold to decide for a mote, which link layer mode to choose, is mainly based on its number of interested children and the duty cycling rate. An implementation of BMRF is realized in Contiki. Our measurements show that BMRF, when using the optimal configuration, results in less radio transmissions, and less energy consumption, and higher packet delivery ratio compared to SMRF, often at the cost of a higher end-to-end delay

This paper presents a new key management protocol for group-based communications in non-hierarchical wireless sensor networks (WSNs), applied on a recently proposed IP-based multicast protocol. Confidentiality, integrity, and authentication are established, using solely symmetric-key-based operations. The protocol features a cloud-based network multicast manager (NMM), which can create, control, and authenticate groups in the WSN, but is not able to derive the actual constructed group key. Three main phases are distinguished in the protocol. First, in the registration phase, the motes register to the group by sending a request to the NMM. Second, the members of the group calculate the shared group key in the key construction phase. For this phase, two different methods are tested. In the unicast approach, the key material is sent to each member individually using unicast messages, and in the multicast approach, a combination of Lagrange interpolation and a multicast packet are used. Finally, in the multicast communication phase, these keys are used to send confidential and authenticated messages. To investigate the impact of the proposed mechanisms on the WSN, the protocol was implemented in ContikiOS and simulated using COOJA, considering different group sizes and multi-hop communication. These simulations show that the multicast approach compared to the unicast approach results in significant smaller delays, is a bit more energy efficient, and requires more or less the same amount of memory for the code.

Many Internet protocol -based multicast protocols for Wireless Sensor Networks (WSNs) rely on the routing tree obtained through the Routing Protocol for Low power and lossy networks, to reach subscribers of a multicast group. WSNs often activate radio duty cycling (RDC) to save energy. Our purpose is to examine the effect of the link layer (LL) transmission mode, on the performance of those multicast protocols in function of RDC and multicast protocol parameters. First, the average packet delay in function of the LL transmission mode and the radio channel check rate (CCR), is investigated for restricted topologies through mathematical modeling and simulation. Both show that LL broadcast produces equal or less delay for reaching its interested children in the first hop compared with LL unicast. However, LL broadcast suffers from higher delays, inversely proportional to the CCR, for subsequent hops. LL unicast's delay per hop increases linearly with the number of interested children. Second, the simulation model is used to study the effect of using a mix of LL unicast and broadcast for different CCRs in randomly generated topologies. Average delay, energy consumption and packet delivery ratio are determined in function of hop distance. The average delay can be optimized by using a mixed mode, in which a threshold determines from how many interested children onwards a LL broadcast should be used. The results indicate which optimal threshold, in function of the CCR, should be chosen to optimize delay and energy consumption.

Smart buildings benefit from IEEE 802.15.4e time slotted channel hopping (TSCH) medium access for creating reliable and power aware wireless sensor and actuator networks (WSANs). As in these networks, sensors are supposed to communicate to each other and with actuators, IPv6 multicast forwarding is seen as a valuable means to reduce traffic. A promising approach to multicast, based on the Routing Protocol for Low Power and Lossy Networks (RPL) is Bidirectional Multicast RPL Forwarding (BMRF). This paper aimed to analyze the performance of BMRF over TSCH. The authors investigated how an adequate TSCH scheduler can help to achieve a requested quality of service (QoS). A theoretical model for the delay and energy consumption of BMRF over TSCH is presented. Next, BMRF’s link layer (LL) unicast and LL broadcast forwarding modes were analyzed on restricted and realistic topologies. On topologies with increased interference, BMRF’s LL broadcast on top of TSCH causes high energy consumption, mainly because of the amount of energy needed to run the schedule, but it significantly improves packet delivery ratio and delay compared to ContikiMAC under the same conditions. In most cases, the LL unicast was found to outperform the LL broadcast, but the latter can be beneficial to certain applications, especially those sensitive to delays.

6LoWPAN, a technology for allowing the deployment of IPv6 on Low Power and Lossy Networks enables interoperability and user-friendliness when establishing applications related to the highly popular trend of Internet of Things. In this paper, we investigate the impact of including a low cost security solution into the communication scheme on latency, power and memory requirements. The measurements demonstrate that this impact is acceptable for most applications. They also show that the impact drastically decreases when the number of transmitted messages decreases or the number of hops increases.