Microwave Engineering Europe - April 2008 - (Page 22)

22 MILITARY/AEROSPACE FOCUS • Collocation issues Traditional narrowband communications equipment covering different bands (for example, VHF equipment for the 30 to 88 MHz band, or UHF equipment for the 225 to 400 MHz band) are inherently physically separated and thus better isolated from foreign inﬂuences. Narrowband antennas have inherent attenuation of out-of-band signals in order of tens of dB, thus signiﬁcantly alleviating the radio front end speciﬁcations. Wideband equipment often requires costly ancillary equipment (preselectors, postselectors, etc.) to resolve the collocation issue. Unfortunately, this ancillary equipment, which is a combination of switches, ﬁlters, ampliﬁers and so on, is normally a narrowband one and has a limited signal (instantaneous) bandwidth. Moreover, collocated wideband communication systems, which by their very nature must be able to operate at any frequency within their band and switch frequency at a moment’s notice, require sophisticated frequency management software to be an integral part of radio networking [2]. It is a huge logistics burden! Digital hardware platforms for software radio The digital hardware platform needed to implement software radio is typically an array of computing devices (GPP, DSP, FPGA, etc.), that can run a computer program implementing all the radio algorithms. The platform must also support RF hardware reconﬁgurations and various control functions. The software communication architecture (JTRS SCA) platform is supposed to allow porting to it every possible waveform (present and future) and operate it properly right away. However, the very concept of designing a platform suitable for future waveforms is problematic “ how do you design a cost effective, reliable platform for something you don’t even know what is it? One approach is to put a lot of computing resources on a board (see, for example, the SDR 3000 Series platforms of SPECTRUM Signal Processing): this is surely a very expensive proposition, not quite competitive in terms of cost, space, and power consumption. Another approach is make it scalable, although it is not very clear how scalability can be maintained in practice. Moreover, many design goals, operational system latency, timing accuracy, memory resources availability and so on, must be clearly stated when specifying the hardware components for the software platform. For example, the printed circuit board (PCB) should be appropriately designed for the highest clock frequency possible and for the most intensive tasks that may ever be run on the platform. Most demanding is probably the timing accuracy issue — timing jitter unavoidable in software platform can be a limiting factor for high complexity waveform implementation. Design and development efforts The main tasks required to design radio systems are summarized in Table 1. Software related tasks are very complex and experience shows that they consume most of the development efforts and resources, and that a signiﬁcant fraction of these efforts is directed to developing and simulating appropriate algorithms. Proper software design and debugging is probably the most lengthy and resource consuming development task, however, its contribution to the product cost depends on the quantity of radios expected to be manufactured. For large quantities, hardware plays a dominant role in the project’s commercial success. Therefore, it is very important to be aware of any trade offs between hardware cost and performance. Pushing hardware capabilities to meet ISWR requirements can easily lead to per unit costs one can hardly afford (average JTRS cost estimates are $127K a unit [2]!) Conclusions Modern communication systems make wide use of embedded computers to implement advanced algorithms for signal processing, networking and data transmission. Still, the radio itself remains a physical, hardware device that basically modulates a carrier wave in order to transmit and receive information in a wireless medium. Many related processes: RF energy generation, ampliﬁcation, conversion into propagating electromagnetic waves at the transmitter side and back into electrical power at the receiver side, removing high power interference and noise and amplifying the useful signal to levels suitable for signal processing at the receiver end — all of them are inherently physical operations. The term “software radio” is therefore somehow misleading in that it ignores physical limitations and the development and production costs incurred when one tries to get too close to the ideal. The radio hardware design engineering expertise is still relevant and ever more challenging in a software radio world. References [1] Antennas, John D. Kraus, McGraw Hill, 1950 [2] The Army’s Bandwidth Bottleneck, a CBO (Congressional Budget Ofﬁce) Study, August 2003 About the author David Lipets is head of group developing modem and RF technologies for tactical and software deﬁned radios in Tadiran Communications Ltd (today part of Elbit Land Systems and C4I). He received BSc in electronics from Tel-Aviv university in 1985 and since then was engaged in several radio design projects. His areas of interest are software deﬁned radio and RF technologies. Reprinted courtesy of RF Designline (www.rfdesignline.com/206902442). Table 1: A summary of the main tasks required to design radio systems. Microwave Engineering Europe ● April 2008 ● www.mwee.com 018-019-020-021-022_MWEE.indd 22 26/03/08 18:07:55 http://www.rfdesignline.com/206902442 http://www.mwee.com

Table of Contents Feed for the Digital Edition of Microwave Engineering Europe - April 2008

Microwave Engineering Europe - April 2008 News Contents Comment Test and Measurement: Comprehensive WiMAX and Wi-Fi Product Design Demands Effective Channel Emulation Military/Aerospace Focus: Hardware Needs Limit Software Radio Interview — Mitsubishi Electric Europe: GaAs Technologies Spanning High-End Space and Radar Through to Cost-Sensitive Handset and LNB Applications How Do You Test ZigBee Transmitters? Advanced Receiver Design Boosts Performance CMOS PAs Pave the Way for One-Chip Phones Products Calendar

Microwave Engineering Europe - April 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.