Virtual Machine for Software Defined Radio: Evaluating the Software VM Approach Riadh Ben Abdallah, Tanguy Risset and Antoine Fraboulet Citi, Insa-Lyon 6, avenue des Arts 69621 Villeurbanne Cedex, France Emails: {riadh.ben-abdallah, tanguy.risset, antoine.fraboulet}@insa-lyon.fr J´ erˆ ome Martin CEA-LETI, MINATEC, 17, rue des Martyrs, 38054 Grenoble Cedex, France Email: jerome.martin@cea.fr Abstract—We study the impact of using a virtual machine for the configuration of radio physical layer protocols on a real hardware platform: the Magali chip. The virtual machine is programmed in software on the ARM processor present on the platform. We evaluate the additional cost of the virtual machine layer on the effective implementation of telecommunication physical layer protocols. The results, obtained using the mixed SystemC/VHDL cycle accurate simulator of the Magali platform, show that, although the proof of concept is valid and functional, extra optimizations, such as additionnal hardware mechanisms, will be necessary to obtain real-time performance. I. I NTRODUCTION Software defined radio is now foreseen as the next tech- nological shift that will drive commercial success for new mobile embedded systems. Automatic and dynamic adaptation to the strongest (or cheapest) radio protocol as well as global minimization of energy consumption over a set of mobile nodes may be reached only with the availability of software reconfiguration of the protocol physical layer. Following the pioneering work of Mitola [1], software defined radio (SDR) has been “de facto” defined as the ability to program the physical layer of the radio protocol used for wireless communication. This does not mean, of course, that the protocol is fully realized in software. High bandwidth telecommunication protocols require hardware components for a real-time implementation. However, the same hardware com- ponents (e.g., a FFT component) are used in several different protocols with different parameters. What must be done in software is the configuration and control of these hardware components. As soon as a SDR platform is programmable, its programs (sometimes called waveform programs) should be easy to develop and reusable. The reasons for that are that i) pro- grammability very quickly leads to intractable complexity and time-consuming debugging process and ii) the vast quantity of hardware mobile platforms makes it impossible to develop one waveform program per hardware platform. This is the basic motivation for the use of a virtual machine for SDR:a dedicated virtual machine available on each hardware platform would be a solution to the “program once run everywhere” ideal scheme for waveform programs. This Radio Virtual Machine concept (RVM) has been pro- posed in various works [2], [3], we have presented in [4] our proposal that includes a specific mechanism for the virtual machine to act on the signal processing data-stream. An important question-mark remains over the overhead induced by the use of a virtual machine for waveform configuration. This overhead can be measured in terms of additional hardware complexity and/or additional software complexity which itself can be divided into run-time performance and compile time issues (e.g., memory used). We propose in the paper to evaluate the practical impact of the use of a virtual machine on an existing platform: the Leti Magali chip [5]. In this work, we investigate a proof of concept to implement a software RVM: we reuse an existing SDR platform without any additional dedicated hardware. Given an existing hardware platform able to execute different waveform programs, such as Magali, we provide answers to the following questions: how much does it cost to add a software virtual machine on this platform? Will the obtained RVM respect real timing con- straints of 3G telecommunication protocols? Our experiments have been done with a port of the Lua virtual machine on the ARM processor present on Magali. These experiments show that our software RVM implementation is approximately 2 to 6 times slower than native implementation which tends to prove that the overhead introduced can be managed and that more optimized VM should be used and investigated. The paper is organized as follows: an overview of existing SDR platforms is presented in section II. Our RVM proposal is then rapidly recalled in section III. Section IV introduces the Magali platform. Implementation choices are then detailed in section V, and section VI describes the experimental tests realized in order to evaluate our RVM implementation. Finally, section VII presents our conclusions. II. EXISTING SDR PLATFORMS SDR offers faster time to market and shortens development cycle of new products. Due to these economic issues, SDR technologies have shown quick advancements during the last few years. An important number of SDR platforms with various architectures have been proposed both by academic research laboratories and by commercial companies. In the following, we briefly present a representative sampling of SDR platforms. 1