Modern cellular networks are complex systems offering a wide range of services and present challenges in detecting anomalous events when they do occur. The networks are engineered for high reliability and, hence, the data from these networks is predominantly normal with a small proportion being anomalous. From an operations perspective, it is important to detect these anomalies in a timely manner in order to mitigate them and preclude the occurrence of major failure events. In telecommunications diverse set of data such as KPIs, logs and alarms are generated to monitor the health and stability of the network element and the services carried over it [1]–[4].

The knowledge of a future link quality for a user equipment (UE) in wireless networks may be used by the application and TCP layer so that user's experience is optimized. The metrics that are relevant are available at the core of the radio link and the application or TCP layers have no knowledge of the wireless metrics. With a signaling protocol to the application/TCP layer and a prediction mechanism of the future link quality the applications will be able to avoid buffer overflows and/or congestion. In this paper we identify metrics that are suitable for application/TCP control, and analyze the performance of the prediction of the metrics. We show that the prediction of wireless metrics can be made with low error (3-10% MAPE) even with high cloud latency. This is a significant result as it states that predictions in the cloud of even short term LTE metrics are possible and that the predictions are fairly accurate to influence the proper operation of applications/TCP in wireless environments.

We describe the design, implementation, and experimental evaluation of a clock recovery algorithm suitable for synchronization over a packet-based metro-area network (MAN) using the Precision Time Protocol (PTP). The algorithm consists of a packet-based phase error estimator and a phase-locked loop (PLL). The loop is augmented with a step detector that detects discontinuities in the packet arrival process due to route or network load changes, and a congestion detector that filters out unusable samples. The PLL requires a number of interesting tradeoffs that are specific for a high noise environment, while the congestion and step detector are non-traditional mechanisms specifically designed to deal with the peculiarities of a packet network.