We explore the feasibility of onboard anomaly detection using artificial neural networks for CubeSat systems and related spacecraft where computing resources are limited. We gather data for training and evaluation using a CubeSat in a laboratory for a scenario where a malfunctioning component affects temperature fluctuations across the control system. This data, published in an open repository, guide the selection of suitable features, neural network architecture, and metrics comprising our anomaly detection algorithm. The precision and recall of the algorithm demonstrate improvements as compared to out-of-limit methods, whereas our open-source implementation for a typical microcontroller exhibits small memory overhead, and hence may coexist with existing control software without introducing new hardware. These features make our solution feasible to deploy on board a CubeSat, and thus on other, more advanced types of satellites.

One of the main application scenario that the upcoming 5G systems are expected to support is Internet of Things (IoT) traffic backhauling. With the exponential increase of devices, it follows the critical challenge of handling the huge data volume that they produce. Terrestrial networks may not be sufficient for that, or not always available. Satellites then come into play, and they can provide the ubiquitous coverage targeted by 5G. On the other side, aggregation can be an efficient strategy to reduce the amount of traffic, and use network resources in an efficient manner. Building on that, in this work, we consider a hybrid IoT-satellite network architecture for collecting message queuing telemetry transport (MQTT) traffic. We design an advanced filter, namely, MQTT message filter aggregator (MQTT-MFA), that performs MQTT topics aggregation in close proximity of MQTT Publishers before data are sent over the satellite link. The developed MQTT-MFA implementation leverages on MQTT bridge, and it is compliant with Mosquitto, an open-source MQTT broker. Extensive simulation results have demonstrated that MQTT-MFA requires only 10% of the TCP/IP overhead generated in a baseline architecture (where the bridge is not set up) and 1/4 number of bytes to transmit the same amount of information, when compression of the data payload is also enabled.

Given that several services can benefit from the adoption of a group communication model, the IETF has specifically standardized the usage of CoAP group communication. However, CoAP responses are still sent in unicast from each single CoAP server to the CoAP client, which results in a substantial traffic load. Such problem becomes more severe in integrated IoT-Satellite networks given the limited bandwidth of the satellite return channel and the large number of IoT devices in a massive MTC (mMTC) scenario. To reduce network traffic overhead in group communication and improve the network responsiveness, this paper proposes an aggregation scheme for the CoAP group communication in combination with Observer pattern and proxying. Results obtained by using the openSAND emulator and CoAPthon library corroborate the merit of our optimization in terms of overhead reduction and delay.

The MQTT application protocol was originally designed for monitoring a oil pipeline through the desert by collecting sensor data via satellite link. Thus, by design MQTT is very suitable for data collection over integrated satelliteterrestrial networks. Leveraging on the MQTT Bridge functionality, in this work we propose a novel architecture with two MQTT Brokers located at the satellite terminal and the satellite gateway. By using the topic pattern option, supported by the bridge, a subscriber can request several topics within a single request. To reduce the amount of traffic exchanged over the satellite return channel, we introduce in the architecture a new entity, namely MQTT message aggregation filter, which aggregates all the MQTT topics matching the topic pattern in the same response.