In this letter, we address the problem of channel classification. Specifically, we consider five channel models also accounting for the incoming fifth generation mobile communication network (5G). The classification problem is formulated in terms of a multiple hypothesis testing problem and solved through a heuristic decision logic relying on sequential binary comparisons. This choice is dictated by the mathematical intractability of probability density functions (PDFs) under some hypotheses which contain transcendental functions. As a consequence, decision rules raising from well-known design criteria based upon data distribution might suffer numerical instability related to the routines used to implement such PDFs. The performance analysis is conducted on simulated data and shows the effectiveness of the proposed strategy in channel classification.

中文翻译：

---

考虑Nakagami-m Shadowing和FTR模型的信道分类方案


在这封信中，我们解决了渠道分类的问题。具体来说，我们考虑了即将到来的第五代移动通信网络（5G）的五种信道模型。分类问题根据多重假设检验问题来表述，并通过依赖于顺序二元比较的启发式决策逻辑来解决。这种选择是由概率密度函数 (PDF) 在某些包含超越函数的假设下的数学难解性决定的。因此，基于数据分布的众所周知的设计标准提出的决策规则可能会遭受与用于实现此类 PDF 的例程相关的数值不稳定。对模拟数据进行性能分析，显示了所提出的策略在信道分类中的有效性。