Wireless distributed computing in cognitive radio networks Dinesh Datla ⇑ , Haris I. Volos, S.M. Hasan, Jeffrey H. Reed, Tamal Bose Wireless @ Virginia Tech, Blacksburg, VA 24061, United States article info Article history: Received 26 January 2011 Received in revised form 5 April 2011 Accepted 6 April 2011 Available online 15 April 2011 Keywords: Distributed computing Cognitive radio networks Cognitive engine Power and energy consumption Workload allocation abstract Individual cognitive radio nodes in an ad-hoc cognitive radio network (CRN) have to per- form complex data processing operations for several purposes, such as situational aware- ness and cognitive engine (CE) decision making. In an implementation point of view, each cognitive radio (CR) may not have the computational and power resources to perform these tasks by itself. In this paper, wireless distributed computing (WDC) is presented as a tech- nology that enables multiple resource-constrained nodes to collaborate in computing com- plex tasks in a distributed manner. This approach has several benefits over the traditional approach of local computing, such as reduced energy and power consumption, reduced burden on the resources of individual nodes, and improved robustness. However, the ben- efits are negated by the communication overhead involved in WDC. This paper demon- strates the application of WDC to CRNs with the help of an example CE processing task. In addition, the paper analyzes the impact of the wireless environment on WDC scalability in homogeneous and heterogeneous environments. The paper also proposes a workload allocation scheme that utilizes a combination of stochastic optimization and decision-tree search approaches. The results show limitations in the scalability of WDC networks, mainly due to the communication overhead involved in sharing raw data pertaining to delegated computational tasks. Ó 2011 Elsevier B.V. All rights reserved. 1. Introduction In ad-hoc cognitive radio networks (CRNs) that are comprised of ideal cognitive radios (iCR) [1], each iCR has to perform complex data processing operations for several purposes, such as situational awareness which includes identification of the characteristics of signals and their sources, and cognitive engine decision making which in- cludes learning and reasoning. The data processing opera- tions include spectrum sensing [2]; signal detection, classification [3] and feature extraction [4]; geo-location to identify location of malicious emitters; knowledge-base and case-base updates and searches; natural language pro- cessing; and computer vision. In the context of a cognitive engine (CE), the following tasks are common: first, esti- mating BER performance based on theoretical BER curves. CEs use those curves to evaluate hundreds of communica- tion configurations and select a few to try over-the-air [5,6]. The BER curves can also be used to initialize an obser- vations database, which stores information about the ob- served effectiveness of various communication methods when employed under different channel conditions, before the CE begins its first operation [7]. Second, estimating confidence intervals of observations in the database [7] re- quires estimation of the Beta distribution CDF which is computationally demanding. Third, CEs that employ case- base reasoning require searching and analyzing past cases [5,8]. Finally, general CE database maintenance requires a lot of computing resources. Such complex computational tasks, with high resource demands and stringent quality-of-service requirements, may cause a processing burden to resource-limited cogni- tive radio (CR) nodes and each CR node may not possess 1570-8705/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.adhoc.2011.04.002 ⇑ Corresponding author. Tel.: +1 785 312 0818. E-mail addresses: ddatla@vt.edu (D. Datla), hvolos@vt.edu (H.I. Volos), hasan@vt.edu (S.M. Hasan), reedjh@vt.edu (J.H. Reed), tbose@vt.edu (T. Bose). Ad Hoc Networks 10 (2012) 845–857 Contents lists available at ScienceDirect Ad Hoc Networks journal homepage: www.elsevier.com/locate/adhoc