There is an increase in vehicle use in Addis Ababa which increases by 5% annually, not onlyrnthis but also the urban arterial roads are of mixed type and there are no automated means thatrnmonitor the road traffic congestion. This creates road traffic congestion which is one of thernmain problems that affects everyone in time, energy and resource management. Road trafficrnestimation is instrumental in developing a more advanced transportation system to avoidrncongestions, trip delay, improper fuel usage, and vehicle use. There are different methods torncollect data and estimate the degree of road traffic congestion such as manual countingrn(observation), global positioning system, inductive loops, using a camera system. The latterrnthree mentioned are expensive in installation and maintenance with coverage limitations. Thisrnthesis focuses on one of the cost-effective alternative methods, which is based on 3G cellularrnhandover data in the case of Addis Ababa's main selected road for the experiment. rnIn this study, we examined this alternative method to estimate the degree of road trafficrncongestion using a simple feedforward backpropagation neural network by considering andrnignoring the altitude parameter. Related research has been done, but this study does not showrnby considering the altitude as a factor. We collect handover data while driving along selectedrnarterial roads in Addis Ababa, and also handover archived data has been used for somernsections of the streets for the experiment. For the congestion estimation data classification,rnAADT(Annual average daily traffic) data has been collected from AAPC(Addis Ababa policerncommission), and LOS(level of services) also used, which is based on the speed of test vehicles.rnThe neural network was then trained and tested using the collected data against the road levelrnclassification done. The results found showed better performance of congestion estimation withrnan accuracy of 92.5% when the feature altitude is considered whereas without considering thisrnvariable, the result found is similar to those of other related research works 82.1 %. It has beenrnshown from the effects that the improvement in the accuracy of estimation increases by 10%.