Universal Platform for Software Defined Radio Ryuji Kohno 1 Masayoshi Abe 2 Noboru Sasho 2 Shinichiro Haruyama 3 Robert M-Zaragoza 3 Elvino Sousa 4 Francis Swarts 3 Pieter Van Rooyen 3 Yukitoshi Sanada 3 Lachlan B. Michael 3 Hamid Amir-Alikhani 5 Veselin Brankovic 5 1: Yokohama National University, Japan 2: Semiconductor Company, Core Technology Network Co., Sony Corporation, Japan 3: Advanced Telecommunication Laboratory, Sony Computer Science Laboratories, Inc., Japan 4: Dept. of Electrical and Computer Engineering, The University of Toronto, Canada 5: Telecommunication Research & Development Stuttgart, Sony International (Europe) GmbH, Germany Abstract In this paper, we propose a novel wireless radio plat- form suitable for software defined radio. This platform is capable of handling multiple radio standards. The pro- posed platform consists of a new broadband radio fre- quency (RF) front-end supported by a reconfigurable dig- ital signal processor (DSP) or field programmable gate array (FPGA) technology. The advantages of our plat- form compared to other SDR architectures are presented in this paper. key words: Software Defined Radio, Universal Platform, Multimode Multiband Receiver, Direct Conversion, Five Port MMIC 1 Introduction After the first technical paper was presented in 1992, soft- ware defined radio (SDR) has been receiving much atten- tion among researchers working on wireless communica- tions [1] [2]. Behind this popularity, there is a conceptual reason and a technical reason [3]. The conceptual reason is that various wireless stan- dards have been established through generations of wire- less communication systems. Even in the same genera- tion, several standards have been created in different re- gions. As an example, the standardization efforts surrounding IMT2000/UMTS have tried to resolve the dispute over what the third generation standard should entail. Despite all of this, it still seems as if three slightly varying CDMA standards will be introduced in the near future. For wireless LAN, not only IEEE standards, but also de facto standards such as Bluetooth have gained wide acceptance among companies all over the world. Thus, a multiband multimode radio system is required to cre- ate a comfortable mobile computing environment and the reconfigurability of SDR is the answer to this problem. The technical reason behind the poularity of the SDR concept is the development of reconfigurable devices for signal processing such as digital signal processors (DSP) and field programmable gate arrays (FPGA). The latest DSP’s operate at speeds up to 1.1 GHz and offer perfor- mance of nearly 9 billion instructions per second. FPGA’s can now provide densities of up to 2 million gates with low power consumption. These numbers are ever improving. [4][5]. Therefore, the real challenges facing SDR are the RF front-end, which is able to use the reconfigurability of the signal processing devices mentioned above and providing multimode and multiband communications. In this pa- per a novel universal platform is proposed for SDR. The proposed platform consists of a new broadband RF front- end followed by reconfigurable devices such as DSPs and FPGAs. The RF front-end employs the direct conversion approach using an original five port MMIC [6]. 2 Direct Conversion MMIC based on Multiport Technology In order to realize a multimode multiband SDR, the RF front-end should be able to support a wide range of fre- quencies and bandwidths. This task may be difficult with conventional RF front-end architectures [7]. The block diagram of a conventional heterodyne re- ceiver is shown in Fig. 1. This architecture requires frequency-dependent passive components such as dielec-