How can we transform our cities into Smart Cities? The race is on, but the rules of the game have not been written yet. Several suggestions are currently being analyzed in several places in the world. Its aim: to achieve a system that radically transforms the way we live in our cities, generating services and resources that do not exist yet. And, above all, reducing costs.
Andrés is working along a street in his city. It’s already dark, and as he walks the node lights keep obediently increasing their intensity as he comes closer to them, and becoming weaker once he moves away. Absorbed in his last-generation cell phone, he is not aware of that; he is also unaware of the fact the the fast communication speed of his mobile device is due to an intelligent data network located in the street, inside the nodes he is walking along.
Somewhere else in the city, in a control room, a municipal technician launches a computational process that will be executed not only in a classical computer from a classical data center, but in the urban computation network that involves all the computational power excess that exists in the tiny microprocessors installed inside the urban nodes. In a few seconds, the launched process is finished, and he starts a new one.
This could seem a science-fiction story, but this is happening right now. Our work group has joint efforts to develop a new urban piece of furniture that includes, from the very beginning, many of the requisites faced in cities all over the world: the urban node.
In the current economic crisis scenario, all cities in the world have seen their investment capacities seriously reduced. Their needs have greatly increased in parallel. Today it is essential to reduce service costs in order to be able to continue offering them, and to achieve that it is essential to reduce consumption.
This new concept could be seen as an active, modular and readjustable element of urban furniture that silently performs several tasks. It gathers information from its surroundings and process it, it supports public telecommunication networks, it generates energy and gives it to the network, it can be used as an information and advertising point, and, if necessary, it may be used as an electricity supply point for already-existing electric vehicles.
Now in the 21st century, more than 140 years have went by since what we could call pre-nodes appeared for the first time: electric lampposts. It’s time to make this old concept move forward and adapt it to modern times, in the framework of new technologies and innovation (innovation should be the fundamental pillar to leave the current crisis behind). In that sense, it was necessary to give it a new name to dissociate it from urban lighting only. Because we should bear in mind that, even if its external shape may resemble that of a lamppost, they are not the same: their function range may include providing light in the streets at night, but this is not essential. We expect its modular design to adapt to the needs of its surroundings wherever it is installed, but perhaps it will not provide light. And anyway, even if it did, it will do so in a smart way, controlling energy consumption and performing real-time analysis of the lighting needs of its surroundings to increase citizen comfort.
The revolution of sensors
One of the aspects that sustain the current technological revolution is the radical decrease in the price of electronic components. This could only be compared to their progressive reduction in size, their integration in microchips and their increase in power. Nowadays, a wide range of sensors are available just for a few tens of euros. This would have been impossible some time ago, as its high price would have discouraged anyone from using just a handful of them in a given place. Thus, an urban node will be able to gather quality data -such as temperature, humidity, atmospheric pressure, several parametres regarding pollution, noise, lighting- at a very low cost, and do it wherever needed.
Their modular design will make it possible to configure nodes according to the particular needs and location possibilites. This will enable to control several factors, as is already being done in pilot tests in cities such as Barcelona: from how full the garbage bins are, in order to coordinate when to empty them, to humidity in soil to initiate watering systems or to raise alerts involving high levels of carbon monoxide in places with high traffic.
From the combined outputs of such simple sensors other factors will be deduced, such as the presence of fire: an increase in temperature in combination with an increase in carbon monoxide levels and a humidity decrease. Such data, if properly processed, could originate a quick intervention system, which would save lives and avoid property damage.
But innovation in sensors is more than all the aforementioned. More complex sensors already exist, and they are capable of directly offering combinations of data: cameras. Low-cost cameras already exist, and they will make a great range of possibilities come true. From automatic processing of images, in such a way that privacy is respected, many information may be obtained: from distance management of car park places -making these data available could reduce the search time, which would in turn bring about a decrease in fuel costs and an improvement in citizen comfort- to traffic management and the detection of traffic accidents.
Our first approach to the urban node design involved an international group of engineering students in their final project. The students, who came from nine different European and American countries, were to face a challenge we had previously defined: urban nodes should produce more energy than the amount they spent while working. Another requisite was to avoid using batteries to store energy, because of their high price and still high maintenance costs.
Clearly enough, these are difficult challenges to tackle, but they are not impossible. Urban nodes will be connected to one of the most extensive public energy networks, that of public lighting, which means they will be connected to the general electrical supply network. At every given time, each node will calculate whether its energy production/consumption is positive or negative, and they will act accordingly, either taking energy from or giving it to the network.
This is how such a relevant problem as energy storage may be solved. In case a positive production rate is achieved in one network node, it will transfer its energy to the municipal supply network -not to that of providing companies- thus making it possible to use this energy somewhere else in the city and reducing the energy needs of the city as a whole and its subsequent economic costs. In this way, taking into consideration that an urban energy network is extremely wide, and that several devices are connected to it (engines, traffic lights, supply pumps…) expenses may be modulated. Only if the network as a whole had a positive rate of energy generation, which would be extremely desirable, would it be necessary to transfer the energy excess to the networks owned by providing companies.
In order to achieve electric energy production, two classic suggestions were put forward: wind energy and solar energy. Obviously, such energy generation vector will not be installed in all nodes (we should remember this is a modular initiative). It would not make much sense to install solar panels in narrow streets flanked by high buildings, where solar energy production is not economically viable, but it would be advisable, from an economic point of view, to install turbines, given that the street configuration makes it easier for air channelling.
Therefore, proper studies should be performed in order to decide which generators should be placed in each place. Obviously, a mediterranean city with a sea shore such as Vilanova i la Geltrú (our city) will be able to benefit from all its sea front, where sunlight and breeze are abundant, whereas in cities inside the country the distribution will need to include other parameters such as orography, searching for high points more appropriate for their installation.
Another subject to be taken into account is the “fuel” supply to the already-existing fleet of electric vehicles. There is a long way to go in this field. Several aspects such as connection standardization, or how to charge for its use, are yet to be determined. Anyway, it has been suggested that the public network for energy supply could be used as a base to provide further supply to such vehicles while they are parked. It is easy to imagine: a given person parks somewhere in the street, identifies himself/herself and, by means of a proper connection site or of induction systems, recharges the vehicle batteries while he/she is at work, goes shopping or goest to the cinema. After using this service, previous identification made it possible to charge the corresponding fee in his/her bank account. Benefits could be distributed among the supplying company, if their energy was used, or go directly to public accounts, given that a proper use of energy was fostered.
Telecommunications. Shall we speak?
One of the main current problems in all cities, and outside them as well, is the location of cell phone base stations. Everybody wants to have cell phone access everywhere, but nobody is happy about there being a cell phone service station close to their home. This is a technical challenge that has not yet been solved. Stations being farther, the emission power of devices must be stronger, which brings about a perverse inverse effect. When we speak or transmit data, the device in our pocket or close to our ear produces more radiation, which is just what we were trying to avoid by pushing service stations further. Having base-stations closer would significantly reduce this problem. But how can this be done without creating a social alarm?
Right now, the fourth generation of mobile communications is already being stablished, and some countries, such as Japan and the United States, are already testing the fifth one. It increases data velocity and the individual capacity of each station. Nevertheless, an additional problem is that the coverage radius decreases. A direct consequence of that will be the need to increase the number of cell phone base stations.
In our proposal we put forward the following suggestion: some urban nodes could keep a small base station inside, which would be low-power and have a limited radius of coverage. Some pico-, nano- or even femto- stations already exist, which tackle the coverage issue in underground car park, inside underground tunnels or in shopping malls. Technology is already available and it is socially accepted.
Moving this concept to the streets would be a similar revolution to the one our good old street lamppost prompted back in its days. There are two possibilities to lighten a street: one of them would involve placing a powerful spotlight at the end of the street, and the other one would be to install many spotlights all along the street. In the former case, a person standing at the end of the street would get too much light (as with light stations), but somebody standing at the other end of the street would hardly get any light: he would lack “light coverage”. Thus, as was the case with lighting, it is advisable to distribute cell phone coverage in small stations (despite technical problems). After all, these are not two different problems but just the same: both light and cell phone waves are electromagnetic radiations: they just differ in wavelength. But we never heard anyone complain about lamppost light.
There is an additional advantage to that system. Each station located in a node would have its own connection to the optic fiber network, which would exponentially increase communication speed. More mini base stations would mean less users for each stations, and so the communication power of each aerial should be less scattered. Each aerial has a limited communication capacity to be shared among all users. This is quite a straightforward equation: the more the aerials, the less the users for each aerial, which means less users among which to share.
Ubiquitous information technology: towards the Internet of Things
One of the reasons that made the European Centre of High-Energy Physics (CERN) famous has little to do with physics. That is the place where the World Wide Web was born, where the embryo of internet as we know it today was developed.
Nowadays, their experiments in particle physics generate such an amount of information that makes it impossible for any computer in the world to process it at the required speed. Not even the most powerful computer in the world would be able to do so. As they knew what would happen beforehand, they have devoted many years to developing strategies that make it possible to process the huge amount of data created in every experiment they perform. The Worldwide LHC Computing Grid (WLCG), a network formed by hundreds of thousands of processors, attempts to take up all the information generated by the Large Hadron Collider at CERN.
At the same time, many centres in the world where low-cost high computing capacity is required use the technology known as Grid Computing. This is a technology that adds up the power of many small computers so that they cooperate in performing a large project. This is a mature technology used on a daily basis.
Our team introduced the use of a low-cost minicomputer in the node design, as an element for its management and control that took its modularity into consideration. The minicomputer to be used is extremely popular nowadays: microcomputer Raspberry pi. We suggested that it should be used along with a microcontroller plaque to manage sensors: Arduino uno.
This makes it possible for nodes to be extremely flexible. Their programmes are not only easy to update in order to increase performance and get rid of possible errors, but it is also possible to change them in order to reconfigure their functionalities. Besides, this would be done on the network, that is to say, remotely. By doing so, its working obsolence would be avoided, and that would make them ready for designs and uses that are not yet created when they are used for the first time.
Apart from all the aforementioned, they make something else possible: using their excess computing capacity collaboratively, on-grid, in order to perform classical processing tasks. That is to say, they may be used as a whole, as if they were a single, huge computer.
Today, the next generation of the super-computer Mare Nostrum is being developed in the BSC (Barcelona Supercomputing Center). New versions using more sophisticated computer versions based on ARM architecture are being developed, similar to the one in our proposal. Using the hundreds of microprocessors installed in the nodes will enable our cities to transform themselves, paradoxically, into real Smart Cities, thus becoming a huge, powerful, flexible and error-tolerant computer. A disruption in one node will not imply a failure in the whole system, just a small reduction in calculation power.
The future is already present
The current goal of the Urban Node Project is the creation -according to specifications that have already been designed and written- a totally functional, real-scale prototype to be evaluated. The final goal is much more ambitious. Fostered by Neapolis, in Vilanova i la Geltrú, and supported by the Universitat Politècnica de Catalunya (Polytechnic University of Catalonia) the aim is not only to design and build a new line of urban equipment, but also to achieve a final, more relevant goal: move forward towards standardization.
Actually, the final shape of the urban node is unimportant but for certain functional and aesthetical reasons. It should be adapted to its surroundings and to the personality of each city. Our students put forward many suggestions in that sense. From an electronic engineering point of view, the final shape is not important either, because plaques and sensors may be replaced by others and its heart (its microprocessor) may be replaced too.
What is really important is the node concept itself, from which some standards may eventually be written up so that several manufacturers may agree on the essentials and foster a new way of industrial development. Only upon agreeing on how, will we be able to manufacture huge amounts of nodes, reduce their costs and promote a new sort of powerful, effective computer programmes.
Clearly enough, our cities should adapt themselves to new challenges, and do it fast. The urban node is a suggestion; perhaps not the only one, but one that would make it possible to solve many existing problems and a great number of future challenges.