Ordered Transmission for Efficient Wireless Autonomy
Yi-Cheng Chen, Brian M. Sadler, Rick S. Blum · 2018 52nd Asilomar Conference on Signals, Systems, and Computers · 2018
Collaborative autonomous systems may rely on wireless communications in dynamic settings. Fundamental to these systems is collaborative decision making, gathering evidence and efficiently producing a joint decision. In many applications fast decision cycles are required and the speed of decision limits the overall system performance. In this paper we develop and analyze a binary hypothesis testing approach using ordered transmission with K \leq N, where N is the number of sensors and K is the number of transmissions to a fusion center. By ordering the local likelihoods the most informative local statistics are transmitted first. Our analysis, using properties of order statistics, characterizes the test and reveals tradeoffs between K and N. This provides a flexible scheme for quick collaborative decision making, with as few as K=1 transmissions, and enables nontraditional tradeoffs by increasing N while keeping K small. The results can be applied to a variety of distributed decision applications.