In this final chapter of this book, we first summarize its major conclusions. We introduced design and operation approaches towards Smart(er) and Cognitive Software Defined Radios, and showed its application in particular stringent cases. Consequently and finally, we end this chapter and book with some closing remarks. Many open challenges still remain to realize the Anything, Anytime, Anywhere aspiration, which definitely continues to be a moving target. Technological innovation is needed to keep up the pace with increasing capacity requirements. Moreover, regulations, economical and even social conditions have to be considered in order to ultimately serve the users optimally.

8.1 “Good Enough” Is “Close Enough to Optimal”

Wireless services have attracted an impressively growing interest over the past decades. This trend is expected to continue as ever more users want to exchange ever more and richer information. Despite great progress in physical layer techniques, we are however running out of wireless capacity. Novel solutions, both technological and regulatory, are therefore needed to enable a sustainable growth in wireless communication. This book focuses on opportunities opened up by the progress towards smart(er) and cognitive radios. In Sect. 8.2 of this chapter we draw the attention to complementary tactics which we believe will also be vital to keep the communication going. The smart(er) and cognitive operation of radios presented in this book, exploit the possibilities offered by recent and emerging standards which enable a more dynamic spectrum access (see Chap. 2). Consequently, radios that exploit these new opportunities to adapt their behavior to the actual circumstances (both in terms of service requirements and communication scenario) can achieve great performance increase. We thereby adapt the taxonomy illustrated in Fig. 8.1. While cognitive radios can especially bring added value when they are built upon flexible hardware (SDRs, reconfigurable radios), this is not the only form of flexibility to be exploited in the wireless environment. Therefore, the case studies discussed in this book do not impose flexible hardware as a prerequisite.

Fig. 8.1
figure 1

Smart(er) to cognitive radio operation

Full size image

We want to acknowledge to the readers that the boundaries between these classes are not easy to draw. While a crystal clear separation would be desired, it is our opinion that the move towards increased radio intelligence will be gradual, rather than step-wise and that this should considered good news. When J. Mitola launched the cognitive radio concept [16], many questions arose if and when this disruptive idea could be implemented in practical systems within regulatory constraints. A gradual and smooth transition is however feasible, as we hope to have shown in this book. “Good enough” is “close enough to optimal” is thereby the main message we want to convey when concluding this book. When searching for solutions for smart(er) and cognitive operation of radios, we follow the same principle. Certainly in a dynamic environment, such as the wireless scene, the search for ultimate optimal operation may take more energy than could be paid off by the actual gains. Moreover, by the time this optimum is found the conditions may have changed already. We have proposed a generic control strategy based on a two phased approach. In a preparation phase, at design time, a representative set of scenarios are analyzed and corresponding sets of optimal settings are determined and stored. In operation, at run time, based on observation of the environment and requirements, the parameter set corresponding to the closes case are applied. Indeed, a “close enough to optimal” and thus “good enough” results can then be achieved. As a next step in the advancement towards true cognitive behavior, systems will then apply learning techniques to further calibrate and improve these heuristics. A main driver of this book is the scarcity of available spectrum. First a case considering licensed spectrum where Secondary Users use the spare wireless capacity without interfering the Primary users more efficiently has been discussed in Chap. 4. Next, application of the approach for coexistence in an unlicensed ISM band was presented in Chap. 5. Also significant energy savings can be achieved through smart(er) operation. This was illustrated in Chap. 6. As a last and most far-going study the gains achieved by adding learning and calibration to the control operation is revealed for a WLAN case in Chap. 7. For future systems, the combined consideration of both spectrum and energy efficiency will be vital to enable a sustainable future wireless Internet.

8.2 Closing Remarks: The End Is Not There nor in Sight

8.2.1 Keep Moving with the Target

We closed a previous book with the statement “This is not the end, its just the beginning” [1]. Following straight engineering logics, the ideas presented in this book building on SDRs and DSA are consequently not anymore the beginning of Software Defined Radio solutions. Yet, this is definitely not the end, nor is the end in sight. In this book we advocate smart(er) and cognitive radios as an answer to the bottleneck that will occur following the ever increasing wireless data traffic within limitations of available spectrum. Recently, the energy consumption of (wireless) communication systems and the impact of the emitted electromagnetic radiation have become a cause for concern. Worldwide, initiatives have been launched and are gaining momentum to radically reduce the overall energy consumption of communication networks and to reduce the carbon footprint of the growing network [130]. If we want to build tomorrows wireless communications systems, we need to minimize the energy consumption and restrict the radiation, but not at the penalty of throughput and intelligent operation. An answer to this challenge showing great potential savings is the approach to restrict wireless communication to short ranges whenever possible. Indeed today many connections, specifically mobile phone calls, connect to a relatively faraway base-station, even if much closer access points (e.g. indoor WLAN) are present. On the longer term, one may even enable 60 GHz-based local wireless access, allowing very directive transmission at data rates that exceed 1 Gbps. This concept based on distributed access in a heterogeneous network environment is illustrated in Fig. 8.2.

Fig. 8.2
figure 2

Future network architectures go distributed for sustainable growth

Full size image

A major part of the radiation and the wasted energy can be avoided and high data throughput is enabled. In order to enable this, multi-standard terminals, which can be realized attractively (both for cost and form factor) when based on reconfigurable radio solutions, will be key. Energy efficient operation of these radios is a main design goal, and cognitive features are clearly desired. The usage of mm-wave communication for short-range access purposes has recently gained interest in standardization [20]. Definitely the 60 GHz spectrum enables wider spectrum still, moreover frequency reuse at very short distances is possible, and thus smaller cells are promoted. Low cost and low power 60 GHz radios will be a vital technology, and are gaining maturity. Directive and adaptive antenna (arrays) both below 10 GHz and for mm wave will be essential to maximize connectivity for minimal energy and radiation. We believe that smart(er) and cognitive radio operation will be key to enable ever more complex wireless communication networks. We hope to have been able to show to our readers that workable solutions can bring significant gain already today. Definitely this is not the end, nor is the end in sight. As (coaches of) enthusiastic researchers we are convinced that great innovation is on its way and that this will enable more wireless capacity for/in less spectrum and energy for many years to come!