It is anticipated that the future cellular networks will consist of an ultra-dense deployment of complex heterogeneous Base Stations (BSs). Consequently, Self-Organizing Networks (SON) features are considered to be inevitable for efficient and reliable management of such a complex network. Given their unfathomable complexity, cellular networks are inherently prone to partial or complete cell outages due to hardware and/or software failures and parameter misconfiguration caused by human error, multivendor incompatibility or operational drift. Forthcoming cellular networks, vis-a-vis 5G are susceptible to even higher cell outage rates due to their higher parametric complexity and also due to potential conflicts among multiple SON functions. These realities pose a major challenge for reliable operation of future ultra-dense cellular networks in cost effective manner. In this paper, we present a stochastic analytical model to analyze the effects of arrival of faults in a cellular network. We exploit Continuous Time Markov Chain (CTMC) with exponential distribution for failures and recovery times to model the reliability behavior of a BS. We leverage the developed model and subsequent analysis to propose an adaptive fault predictive framework. The proposed fault prediction framework can adapt the CTMC model by dynamically learning from past database of failures, and hence can reduce network recovery time thereby improving its reliability. Numerical results from three case studies, representing different types of network, are evaluated to demonstrate the applicability of the proposed analytical model.
TopI. Introduction
The envisioned fifth Generation (5G) Cellular Networks are expected to achieve 1000 times capacity gain mainly through extreme network densification (Figure 1) (Imran, Zoha & Abu-Dayya, 2014). Moreover, with each successive generation of cellular networks, complexity of BS has continued to increase i.e. a typical 2G cell has 500 parameters to optimally configure and maintain; 3G cell has 1000; and 4G has roughly 1500 parameters. Without intervening measures, same complexity growth trend is expected for 5G (Imran et al., 2014). To efficiently manage such an ultra-dense, complex, heterogeneous cellular networks, the paradigm of SON has recently been
Figure 1. Ultra dense heterogeneous complex cellular network
investigated heavily to automate network configuration and management tasks (Hämäläinen, Sanneck & Sartori, 2012). Realizing the importance of SON as a key enabler for future cellular networks, a number of SON use cases have already been standardized by 3GPP (European Telecommunications Standards Institute, 2014). There is a general consensus that SON will not be a luxury but a necessity in 5G networks (Østerbø & Grøndalen, 2012). SON functions, classified as
- 1.
Self-Configuring,
- 2.
Self-Optimizing, and
- 3.
Self-Healing.
They operate by reconfiguring a number of network parameters. For a thorough review of state of the art SON function see Aliu, Imran, Imran & Evans (2013).
Cellular networks are inherently subject to cell outages caused by either BS hardware and/or software malfunctions or misconfiguration of several hundred cell parameters during routine network operation. Forthcoming cellular networks are susceptible to even higher cell outages rates as the multiple SON functions may be subjected to large number of potential conflicts when operated concurrently in a system. Given the parametric overlap as well as coupling among the objectives of different SON functions, it has been shown in Lateef, Imran & Abu-Dayya (2013) that large number of conflicts are possible among SON functions if no self-coordination mechanism is employed. At times, such conflicts can actually degrade networks performance instead of improving it. For example capacity and coverage optimization SON function might try to improve coverage by increasing transmission power. This may conflict with energy efficiency SON. In summary, as identified in Lateef et al. (2013) in an uncoordinated SON, a variety of conflicts may occur when:
- 1.
Two or more SON functions try to modify the same network configuration parameter.
- 2.
A SON function is triggered by an input parameter whose value is dependent upon some other network parameters.
- 3.
There is a change in network conditions by impromptu addition or removal of relay, eNB or Home eNB (HeNB).
- 4.
Different SON function actions try to alter the same KPI of a cell, while adjusting different network configuration parameters 5) A SON function computes new parameter configuration values based on outdated measurements.
- 5.
There is a logical dependency among the objectives of SON functions.