A comparison of battlefield carrier sense multiple access (CSMA) networks with theoretical CSMA network analysis

Tactical network designers are faced with difficult protocol choices. A common choice available is a carrier sense multiple access (CSMA) network. For example, in designing the US Army´s Lower Tactical Internet (TI), Enhanced Position Location Reporting System (EPLRS) CSMA is often considered for alternative uses. There are many benefits to using CSMA; it generally requires little planning and is efficient for dissemination of multicast or broadcast traffic. Its drawback is network contention leading to access delays and collisions. It also does not guarantee reliability. When making decisions on when and how to employ CSMA style networks, it is easy to rely on theoretical CSMA analysis to estimate performance for different loading levels. However, the assumptions made by these analyses are often inconsistent with the actual implementation of the radio systems and with the nature of battlefield networks. This paper, using EPLRS CSMA networks in the TI, identifies these differences, evaluates the potential impact that using incorrect assumptions may have on network design, and discusses network simulation as an alternative for quantifying these differences