Content
Content
1. INTRODUCTION
Moore’s Law has long offered a roadmap for advances in computing power. In 1965, Gordon Moore predicted that computer processing power should double approximately every 12 months, eventually amending this prediction to 18 months. Developments in computer chip design have kept up with the current version of Moore’s Law, but this pace cannot continue indefinitely. As we will describe below, the laws of physics impose a physical limit on how much processing power can be achieved with a silicon chip. Similarly, the ability of medicine to treat diseases more effectively is limited by the ability of current biotechnology to interface with the processes of the body. We do not yet have an effective way for silicon based computers to interact directly with the chemical processes of the human body in order to diagnose and treat illnesses. Biological computing has the potential to solve both problems. This paper sets the stage for considering this topic by examining the limitations of the silicon-based computing paradigm and discussing the other alternative paradigms. It then focuses on biological computing based on DNA(Deoxy-ribo Nucleic Acid) and considers the benefits and possible problems of this radically different form of computing.
1 Dept. Of Computer Science & Engineering, SIST
2. Limitations of Current Computing Technology
The processing power of current computing technology ie silicon-based computing is possible up to the point of the limitations imposed on it by the laws of physics. But there are additional problems with silicon-based technology that make finding viable alternatives even more imperative. First, let’s look at the limitations imposed by physics. The ‘processing power’ of computers is usually measured by speed. Here speed means how fast circuits can move information from A to B and how fast the information can be processed once it gets to B. The traditional computing design paradigm focuses on decreasing the distance that information (in the form of electrical signals) have to travel; in other words, shortening the distance between A and B. This has meant packing more and more processing elements or transistors into the central processing chip of the computer. Each of these transistors is essentially a tiny binary on/off switch. Today, this packing of transistors has reached an amazing level of density. For instance, a common Pentium IV chip packs 55 million transistors in a space the size of a dime. By relentlessly pursuing this miniaturization strategy, computing technology has advanced rapidly in a comparatively short amount of time. To put the advances in perspective, compare the Pentium chip equipped desktop with the ENIAC computer of the 1940s. That device, with its 17,000 vacuum tubes (the transistor’s predecessor) weighed 30 tons and filled an entire room. Yet, its processing power was less than one hundred thousandth that of the Pentium. As relentless as this progress has been thus far, it is dependent upon continuing to find new means to shrink transistors so as to fit ever more of them onto a chip. In the coming years, transistors will have decreased in size to such a great extent that the only way to make them smaller is to construct them out of individual atoms or small groupings of atoms. Unfortunately, quantum effects of physics operating on that size scale will prevent the effective transmission of signals. For example, the Heisenberg Uncertainty Principle states that particles
2 Dept. Of Computer Science & Engineering, SIST
(such as the electrons that make up the information signals that flow through computers) can exhibit the strange behavior of being in other places than where they should be. Researchers can never be completely certain where these particles are at any given moment. This means that electrons, which should ideally be speeding down the atomic-scale circuit pathways in future silicon computers, might be someplace else along with the information they were assigned to carry! Given that the circuit pathways of computers must be able to reliably transmit information; this quantum effect that emerges at the atomic scale is clearly a problem. Thus, there is a lower size limit that the laws of physics impose upon silicon-based computers. This limit may be reached as soon as 10-15 years from now. Reaching this upper limit to the processing power of silicon-based computers could be very problematic for businesses. Though current and future computers will have sufficient power for handling many day to- day operations, e.g. email, future business software needed for activities such as hyper-realistic strategic planning simulations, mapping more efficient airline routes, etc. will ultimately require more processing power than silicon-based computing will be able to provide. Software designers are always in the process of designing software that forces hardware designers to keep adding more power to their creations. As this continues, it will force silicon-based computers up to their design limits. There are two other problems with silicon-based computers. First, the components out of which computer processing chips are made are toxic, e.g. arsenic, and therefore present challenges in both fabrication and disposal. Second, siliconbased computers are not very energy efficient. They waste a great deal of energy in the form of the heat that they generate and the energy they consume. With these limitations in mind, let’s look at some alternatives to the current computing paradigm.
3 Dept. Of Computer Science & Engineering, SIST
3. Alternatives to Silicon based Computing
Researchers have pursued a number of alternatives to silicon-based computing.These have included biological, optical, and quantum computing. Optical computing is based on replacing with light pulses instead of the electrical signals that carry information through silicon-based computers. Information can travel at the speed of light through information pathways in the optical computing scheme far faster than the commonly used pathways in silicon-based computers. Quantum computers, by contrast, use quantum states of subatomic particles to represent information values. While information is essentially limited to binary (on-off) values in silicon and optical computers, quantum computing permits each information element to carry multiple values simultaneously. Quantum computing also exploits the phenomenon of quantum entanglement, which essentially allows any given information state to exist in two locations simultaneously. This allows the equivalent of instantaneous transfers of information from location to location. It should be noted that a great deal of development works remains to be done before either optical computing or quantum computing can produce practical devices for commercial use. Molecule cascade computing is the newest area in the development of alternatives to traditional computing. This technique is based on forming circuits by creating a precise pattern of carbon monoxide molecules on a copper surface. By nudging a single molecule, it has been possible to cause a cascade of molecules, much like toppling dominoes. Different molecules can represent the 1s and 0s of binary information, making possible calculations. While this technique may make possible circuits hundreds of thousands of times smaller than those used today, it shares with the other alternatives the fact that a number of problems must be solved for it to ever be suitable for practical applications.
4 Dept. Of Computer Science & Engineering, SIST
4. Biological Computing
Biological computing is the use of living organisms or their component parts to perform computing operations or operations associated with computing, e.g. storage. The various forms of biological computing take a different route than those used by quantum or optical computing to overcome the limitations to performance that silicon-based computers face. Rather than focusing on increasing the speed of individual computing operations, biological computing focuses on the use of massive parallelism, or the allocation of tiny portions of a computing task to many different processing elements.Each element in and of itself cannot perform its task quickly, but the fact that there is an incredibly huge number of such elements, each performing a small task, means that the processing operation can be performed far more quickly. Silicon-based computers have used massively parallel processing but will never be capable of the level of massively parallel processing that biological computers can demonstrate. The biological nature of the operation of biological computing also makes it uniquely suited to controlling processes that would require an interface between other biological processes and human technology. The table below compares biological computing with silicon-based computing in several areas, including the component materials, processing scheme, maximum operations per second, presence of toxic components, and energy efficiency. Research in the field of biological computing is focusing on the development of a number of different, though related, forms of technology. While all of these forms share the biological components listed above, they share little else and are best thought of as distant cousins. Some of these technologies will likely be applicable to a variety of problems, while others are best thought of as tools suited to specific purposes.
5 Dept. Of Computer Science & Engineering, SIST
5. DNA Computing
INTRODUCTION TO DNA DNA means DeoxyriboNucleic Acid. The complete set of instructions for making an organism is called its GENOME. It contains the master blueprint for all cellular structures and activities for the lifetime of the cell or organism. Found in every nucleus of a persons many trillions of cells, the human genomes consists of tightly coiled threads of deoxyribonucleic acid (DNA) and associated protein molecules, organized into structures called chromosomes. In humans, as in other higher organisms, a DNA molecule consists of two strands that wrap around each other to resemble a twisted ladder whose sides, made of sugar and phosphate molecules are connected by rings of nitrogen-containing chemicals called bases. Each strand is a linear arrangement of repeating similar units called nucleotides. A nucleotide is composed of one sugar, one phosphate, and a nitrogenous base.
6 Dept. Of Computer Science & Engineering, SIST
7 Dept. Of Computer Science & Engineering, SIST
For users who already have an Internet access and an audio-capable PC. This scenario can take advantage of integration with other Internet services such as World Wide Web, instant messaging, e-mail, etc.
PC to telephone or telephone to PC:
Figure 2 PC to Phone or Phone to PC Scenario In this scenario, PC-callers may reach also the PSTN users. A gateway converting the Internet call into a PSTN call has to be used. Traditional telephone users also can make a call to a PC going through the gateway that connects the IP network with PSTN.
8 Dept. Of Computer Science & Engineering, SIST
Telephone to telephone:
Figure 3 Phone-to-Phone Scenarios The IP network can be a dedicated backbone to connect PSTN. Gateways should connect PSTN to the IP network.
6. BASIC SYSTEM COMPONENTS OF VoIP
There are three major system components to VoIP technology: clients, servers, and gateways.
9 Dept. Of Computer Science & Engineering, SIST
Clients: The client comes in two basic forms. It is either a suite of software running on a user’s PC that allows the user, through a GUI, to set-up and clear voice calls, encode, packetize and transmit outbound voice information from the user’s microphone and receive, decode and play inbound voice information through the user’s speaker or headsets. The other type of client, known as a ‘virtual’ client, does not have a direct user interface, but resides in gateways and provides an interface for users of POTS. Servers: In order for IP Telephony to work and to be viable as a commercial enterprise, a wide range of complex database operations, both realtime and non-real-time, must occur transparently to the user. Such applications include user validation, rating, accounting, billing, revenue collection, revenue distribution, routing (least cost, least latency or other algorithms), management of the overall service, downloading of clients, fulfillment of service, registration of users, directory services, and more.
Gateways: VoIP technology allows voice calls originated and terminated at standard telephones supported by the PSTN to be conveyed over IP networks. VoIP "gateways" provide the bridge between the local PSTN and the IP network for both the originating and terminating sides of a call. To originate a call, the calling party will access the nearest gateway either by a direct connection or by placing a call over the local PSTN and entering the desired destination phone number.
10 Dept. Of Computer Science & Engineering, SIST
The VoIP technology translates the destination telephone number into the data network address (IP address) associated with a corresponding terminating gateway nearest to the destination number. Using the appropriate protocol and packet transmission over the IP network, the terminating gateway will then initiate a call to the destination phone number over the local PSTN to completely establish end-to-end two-way communications. Despite the additional connections required, the overall call set-up time is not significantly longer than with a call fully supported by the PSTN. The gateways must employ a common protocol - for example, the H.323 or SIP or a proprietary protocol - to support standard telephony signaling. The gateways emulate the functions of the PSTN in responding to the telephone's on-hook or off-hook state, receiving or generating DTMF digits and receiving or generating call progress tones. Recognized signals are interpreted and mapped to the appropriate message for relay to the communicating gateway in order to support call set-up, maintenance, billing and call tear down.
7. BENEFITS OF VOIP
11 Dept. Of Computer Science & Engineering, SIST
Voice communications will certainly remain as basic form of interaction among people. A simple replacement of PSTN is hard to implement in short term. The immediate goal for many VoIP service providers is to reproduce existing telephone capabilities at a significantly lower cost and offer a quality of service competitive to PSTN. In general, the benefits of VoIP technology can be the following:
1. Low cost: By avoiding traditional telephony access charges and settlement, a caller can significantly reduce the cost of long distance calls. Although the cost reduction is somewhat related to future regulations, VoIP certainly adds an alternate option to existing PSTN services. Only one physical network is required to deal with both voice/fax and data traffic instead of two physical networks. Having only one physical network has the advantages that lower physical equipment cost, lower maintenance costs. 2. Network efficiency: Packetized voice offers much higher bandwidth efficiency than circuit-switched voice because it does not take up any bandwidth in listening mode or during pauses in a conversation. It is a big saving when we consider a significant part of a conversation is silence. The network efficiency can also be improved by removing the redundancy in certain speech patterns. If we were to use the same 64 Kbps Pulse Code Modulation (PCM) digital-voice encoding method in both technologies, we would see that bandwidth consumption of packetized voice is only a fraction of the consumption of circuit-switched voice. The packetized voice can take advantage of the latest voice-compression algorithms to improve efficiency. 3. Simplification and consolidation:
12 Dept. Of Computer Science & Engineering, SIST
An integrated infrastructure that supports all forms of communication allows more standardization and reduces the total equipment and management cost. The combined infrastructure could support bandwidth optimization and a fault tolerant design. Universal use of the IP protocols for all applications reduces both complexity and more flexibility. Directory services and security services could be more easily shared. 4. Single network infrastructure: When installing VoIP in the office only a single cable is required to the desk, for both telephone and data eliminating separate telephone wiring. 5. VoIP uses "soft" switching: VoIP uses "soft" switching which eliminates most of the legacy PBX equipment. Reducing the cost of installing a communications infra-structure and the maintenance cost once installed. 6. Simple upgrade path: The VoIP PBX technology is software based. It is easier to expand, upgrade and maintain than its traditional telephony counterparts. 7. Bandwidth efficiency: VoIP can compress more voice calls into available bandwidth than legacy telephony. IP Telephony helps to eliminate wasted bandwidth by not transporting the 60% of normal speech which is silence.
IP - the underlying protocol - is supported by most platforms and is independent of the transport protocol used.
8. DEVELOPMENT CHALLENGES
13 Dept. Of Computer Science & Engineering, SIST
The goal of VoIP developers is to add telephone calling capabilities to IPbased networks and interconnect these to traditional public telephone network and to private voice networks maintaining current voice quality standards and preserve the features everyone expects from the telephone. We can summarize the technical challenges as the following. 1. Quality of Service (QoS): The voice quality should be comparable to what is available using the PSTN, even over networks of varying levels of QoS. The following factors decide the VoIP quality: 2. Packet loss: In order to operate a multi-service packet based network at a commercially viable load level, random packet loss is inevitable. This is particularly true with communications over the Internet where traffic profiles are highly unpredictable and the competitive nature of the business drives corporations to load their networks to the maximum. Packetizing voice codec are becoming better at reducing sensitivity to packet loss. The main approaches are smaller packet sizes, interpolation (algorithmic regeneration of lost sound), and a technique where a low-bit-rate sample of each voice packet is appended to the subsequent packet. Through these techniques, and at some cost of bandwidth efficiency, good sound quality can be maintained even in relatively high packet loss scenarios. As techniques for reducing sensitivity to packet loss improve, so a new opportunity for the achievement of even greater efficiencies is presented. This refers to the suppression of the transmission of voice packets whose loss is determined by the encoder to be below a threshold of tolerability at the decoder.
14 Dept. Of Computer Science & Engineering, SIST
This is particularly attractive in the packet based networking world where statistical multiplexing favors the reuse of freed-up bandwidth. 3. Delay: Two problems that result from high end-to-end delay in a voice network is echo and talker overlap. Echo becomes a problem when the round-trip delay is more than 50 milliseconds. Since echo is perceived as a significant quality problem, VoIP systems must address the need for echo control and implement some means of echo cancellation. Talker overlap (the problem of one caller stepping on the other talker’s speech) becomes significant if the one-way delay becomes greater than 250 milliseconds. The end-to-end delay budget is therefore the major constraint and driving requirement for reducing delay through a packet network. Propagation delay (the time taken for the information wave-front to travel a given distance through a given media), jitter buffering, packetization, analog to digital encoding and digital to analog decoding delays are responsible for most of the overall delay. Service and wait time through the switching and transmission elements of the network may be considered trivial given the small packet sizes and relatively wide bandwidths prevalent on the Internet. It is generally true that when considering the achievable quality of a given service, the overall geographic distance traveled by a call is far more important than the complexity of its routing, (i.e. the number of intermediary nodes or "hop-count"). 4. Jitter: Jitter is the variation in inter-packet arrival time as introduced by the variable transmission delay over the network. Removing jitter requires collecting packets and holding them long enough to allow the slowest packets to arrive in time to be played in the correct sequence, which causes additional delay. The jitter buffers add delay, which is used to remove the packet delay variation that each packet is subjected to as it transits the packet network.
15 Dept. Of Computer Science & Engineering, SIST
5. Overhead: Each packet carries a header of various sizes that contains identification and routing information. This information, necessary for the handling of each packet, constitutes ‘overhead’ not present with circuit switching techniques. Small packet size is important with real-time transmissions since packet size contributes directly to delay and the smaller the packet size, the less sensitive a given transmission would be to packet loss. Various new techniques such as header compression are evolving to reduce the packet overhead in IP networks. It is likely that packet based networks, of one form or another, will eventually approach the efficiency, with respect to overhead, of circuit-based networks. 6. User friendly design: The user need not know what technology is being used for the call. He should be able to use the telephone as he does right now. 7. Easy configuration: An easy to use management interface is needed to configure the equipment. A variety of parameters and options such as telephony protocols, compressing algorithm selections, dialing plans, access controls, PSTN fall back features, port arrangement etc. are to be taken care of. 8. Addressing/Directories: Telephone numbers and IP addresses need to be managed in a way that it is transparent to the user. PCs that are used for voice calls may need telephone numbers. IP enabled telephones IP addresses or an access to one via DHCP protocols and Internet directory services will need to be extended to include mappings between the two types of addresses. 9. Security issues:
16 Dept. Of Computer Science & Engineering, SIST
VOIP networks introduce some new risks to carriers and their customers, risks that are not yet fully appreciated. Responding to these threats requires some specific techniques, comprehensive, multi-layer security policies, and firewalls that can handle the special latency and performance requirements of VoIP. It is important to remember that a VoIP network is an IP network. Any VoIP device is an IP device, and it's therefore vulnerable to the same types of attacks as any other IP device. In addition, a VoIP network will almost always have non-VoIP devices attached to it and be connected to other mission-critical networks. Every IP network, regardless of how private it is, eventually winds up connected to the global Internet. Even if it is not possible to directly route a packet from the "private" network onto the Internet, it is extremely likely that some host on the "private" network will also be connected to a less private network. Compromising this host provides an attacker with a gateway into the presumed secure private network. It's important, therefore, to secure all IP networks, but VoIP networks have special security requirements. Specific techniques, comprehensive policies, and VoIP-capable firewalls are needed to do the job right.
9. VOIP APPLICATIONS
17 Dept. Of Computer Science & Engineering, SIST
Cross-platform connections Some VoIP products, including Skype and Gizmo Project, run on Windows and Linux, while Apple’s iChat AV runs only on OS X. Skype, SightSpeed, Gizmo Project, and iChat AV allow you to host either multiparty voice or videoconference calls. Unlike expensive high-end conferencing systems designed for large businesses, which are often connected to a telephone system, these simple desktop VoIP apps can make conferencing easier —and more affordable. All of these applications allow you to call other Internet users for free. But if you want to call somebody using his or her telephone number, as permitted by Skype, Gizmo Project, and the Wengo plug-in, you’ll pay a basic, per-minute fee. At this writing, neither iChat AV nor Sigh Speed permits computer-to-phone calling. Advanced features cost money While you can make basic calls for free, more-advanced features will cost you. For instance, Skype’s voice-mail feature carries a small monthly charge. Obtaining a permanent phone number from Skype (called a SkypeIn number) involves an additional fee. Also for a fee, Gizmo Project allows you to forward your incoming calls to another telephone, such as your cell phone, and SightSpeed offers extended conferencing and video-messaging features for paid subscribers. iChat AV users can’t call traditional phone numbers, but they can call each other, using securely encrypted audio channels on the Internet if all participants are .Mac subscribers. Once you become accustomed to a desktop VoIP tool, you may find that VoIP calling becomes a part of your daily routine. After all, it’s a lot easier to dial a Skype buddy by double-clicking on a name than it is to look up a number in Address Book and manually punch it in on your telephone’s keypad. If you’re into multiplayer Internet games, using a tool like Skype to keep in touch with your teammates is nice, as it relieves you from having to type text-chat
18 Dept. Of Computer Science & Engineering, SIST
messages during the game. And if you have relatives in other countries, talking to them over the Internet will cost you a lot less than placing international longdistance calls.
VoIP in Military Applications Military organizations located worldwide are currently transitioning their telephony infrastructure from legacy TDM to Next Generation Networks (NGN) based on VoIP technology. The reasons for the migration taking place within the military have many similarities to those of the migration in the commercial telephony space. Some of these similarities include lower OPEX resulting from having a single consolidated network for data and telephony, and the ability to deliver new services quickly. However, deployments within the military can have additional specific benefits when moving to VoIP. Voice over IP, as its name implies, traverses over IP networks which, when designed correctly, can be more resilient than TDM networks. IP networks are easier to deploy and manage compared to their older TDM counterparts. In the past, significant investments were made in legacy telephony equipment by military organizations. In most cases, this investment included military purpose built TDM equipment. As a result, the process of migration to NGN will be gradual. It may take many years before full end-to-end VoIP communication can be realized - where all handsets are SIP based and TDM trunks are eliminated.
Mobile VoIP Today, as cost efficient communications increase in demand, Mobile VoIP technologies that turn a mobile device into a SIP client and use a
19 Dept. Of Computer Science & Engineering, SIST
data network for sending and receiving communications, have become increasingly important to users. The mobile VoIP market is expected to be worth $32.2 billion by 2013 and by 2019, half of all mobile calls will be made over allIP networks, according to recent industry reports. Mobile VoIP provider REVE Systems offers operators support for the shift to mobile VoIP with their iTel Mobile Dialer Express, a mobile application that makes it possible to use VoIP via any mobile phone and that can be branded by operators. iTel Mobile Dialer Express supports GPRS, Wi-Fi and Bluetooth for Internet connectivity and can run on any phone on Symbian or Windows Mobile 5 and 6 platforms.
10. IS VOIP THE FUTURE OF TELECOMMUNICATIONS?
20 Dept. Of Computer Science & Engineering, SIST
VoIP means that the technology used to send data over the Internet is now being used to transmit voice as well. The technology is known as packet switching. Instead of establishing a dedicated connection between two devices (computers, telephones, etc.) and sending the message "in one piece", this technology divides the message into smaller fragments, called 'packets'. These packets are transmitted separately over a decentralized network and when they reach the final destination, they're reassembled into the original message. VoIP allows a much higher volume of telecommunications traffic to flow at much higher speeds than traditional circuits do, and at a significantly lower cost. VoIP networks are significantly less capital intensive to construct and much less expensive to maintain and upgrade than legacy networks (traditional circuit-switched networks). Since VoIP networks are based on Internet protocol, they can seamlessly and cost-effectively interface with the high technology, productivity-enhancing services shaping today's business landscape. These networks can seamlessly interface with web-based services such as virtual portals, interactive voice response (IVR), and unified messaging packages, integrating data, fax, voice, and video into one communications platform that can interconnect with the existing telecommunications infrastructure. Industry experts see VoIP as a tool that will become the standard platform for the international calling market. It is strongly believed that the profit realization VoIP will trigger in the global telecommunications industry will dwarf the impact of the now ancient "digital revolution". As with any promising new technology, a myriad of companies are trying to climb aboard the VoIP bandwagon. Currently, however, the industry is characterized by a high degree of confusion. Most companies, including large resource-rich national and international telecommunications carriers, are experiencing enormous difficulty in building effective international VoIP
21 Dept. Of Computer Science & Engineering, SIST
networks. They are unable to harness the power of VoIP or effectively communicate the benefits of VoIP to their customers. We mustn’t ignore the problem of scalability. The system has to be designed so that it can grow. And each segment within the system must be able to grow. Creating an architecture that can handle billions of minutes of use per month requires a solution with high call processing capabilities. If we are looking at a global solution, we have to start from the beginning with a global approach. And that’s one of the reasons why a fully implemented solution won’t be available tomorrow.
11. CONLUSION
22 Dept. Of Computer Science & Engineering, SIST
Data traffic has traditionally been forced to fit onto the voice network (using modems, for example). The Internet has created an opportunity to reverse this integration strategy – voice and facsimile can now be carried over IP networks, with the integration of video and other multimedia applications close behind. The Internet and its underlying TCP/IP protocol suite have become the driving force for new technologies, with the unique challenges of real-time voice being the latest in a series of developments. Telephony over the Internet cannot make compromise in voice quality, reliability, scalability, and manageability. Future extensions will include innovative new solutions including conference bridging, voice/data synchronization, combined real-time and message-based services, text-to- speech conversion and voice response systems. The market for VoIP products is established and is beginning its rapid growth phase. Producers in this market must look for ways to improve their time-to market if they wish to be market leaders. Buying and integrating predefined and pre-tested software (instead of custom building everything) is one of the options. Significant benefits of the “buy vs. build “ approach include reduced development time, simplified product integration, lower costs, offloading of standard compliance issues, and fewer risks. Software that is known to conform to standards, has built-in accommodation for difference in national telephone systems, has already been optimized for performance and reliability, and has “plug and play” capabilities can eliminate many very time-consuming development tasks.
12. REFERENCES
23 Dept. Of Computer Science & Engineering, SIST
1. 2. 3. 4. 5. 6.
Computer Networks by Andrew S.Tanenbaum Internetworking with TCP/IP by Douglas E.comer www.iec.org.com www.telogy.com www.rad.com www.mailto:[email protected]
24 Dept. Of Computer Science & Engineering, SIST
Moore’s Law has long offered a roadmap for advances in computing power. In 1965, Gordon Moore predicted that computer processing power should double approximately every 12 months, eventually amending this prediction to 18 months. Developments in computer chip design have kept up with the current version of Moore’s Law, but this pace cannot continue indefinitely. As we will describe below, the laws of physics impose a physical limit on how much processing power can be achieved with a silicon chip. Similarly, the ability of medicine to treat diseases more effectively is limited by the ability of current biotechnology to interface with the processes of the body. We do not yet have an effective way for silicon based computers to interact directly with the chemical processes of the human body in order to diagnose and treat illnesses. Biological computing has the potential to solve both problems. This paper sets the stage for considering this topic by examining the limitations of the silicon-based computing paradigm and discussing the other alternative paradigms. It then focuses on biological computing based on DNA(Deoxy-ribo Nucleic Acid) and considers the benefits and possible problems of this radically different form of computing.
1 Dept. Of Computer Science & Engineering, SIST
2. Limitations of Current Computing Technology
The processing power of current computing technology ie silicon-based computing is possible up to the point of the limitations imposed on it by the laws of physics. But there are additional problems with silicon-based technology that make finding viable alternatives even more imperative. First, let’s look at the limitations imposed by physics. The ‘processing power’ of computers is usually measured by speed. Here speed means how fast circuits can move information from A to B and how fast the information can be processed once it gets to B. The traditional computing design paradigm focuses on decreasing the distance that information (in the form of electrical signals) have to travel; in other words, shortening the distance between A and B. This has meant packing more and more processing elements or transistors into the central processing chip of the computer. Each of these transistors is essentially a tiny binary on/off switch. Today, this packing of transistors has reached an amazing level of density. For instance, a common Pentium IV chip packs 55 million transistors in a space the size of a dime. By relentlessly pursuing this miniaturization strategy, computing technology has advanced rapidly in a comparatively short amount of time. To put the advances in perspective, compare the Pentium chip equipped desktop with the ENIAC computer of the 1940s. That device, with its 17,000 vacuum tubes (the transistor’s predecessor) weighed 30 tons and filled an entire room. Yet, its processing power was less than one hundred thousandth that of the Pentium. As relentless as this progress has been thus far, it is dependent upon continuing to find new means to shrink transistors so as to fit ever more of them onto a chip. In the coming years, transistors will have decreased in size to such a great extent that the only way to make them smaller is to construct them out of individual atoms or small groupings of atoms. Unfortunately, quantum effects of physics operating on that size scale will prevent the effective transmission of signals. For example, the Heisenberg Uncertainty Principle states that particles
2 Dept. Of Computer Science & Engineering, SIST
(such as the electrons that make up the information signals that flow through computers) can exhibit the strange behavior of being in other places than where they should be. Researchers can never be completely certain where these particles are at any given moment. This means that electrons, which should ideally be speeding down the atomic-scale circuit pathways in future silicon computers, might be someplace else along with the information they were assigned to carry! Given that the circuit pathways of computers must be able to reliably transmit information; this quantum effect that emerges at the atomic scale is clearly a problem. Thus, there is a lower size limit that the laws of physics impose upon silicon-based computers. This limit may be reached as soon as 10-15 years from now. Reaching this upper limit to the processing power of silicon-based computers could be very problematic for businesses. Though current and future computers will have sufficient power for handling many day to- day operations, e.g. email, future business software needed for activities such as hyper-realistic strategic planning simulations, mapping more efficient airline routes, etc. will ultimately require more processing power than silicon-based computing will be able to provide. Software designers are always in the process of designing software that forces hardware designers to keep adding more power to their creations. As this continues, it will force silicon-based computers up to their design limits. There are two other problems with silicon-based computers. First, the components out of which computer processing chips are made are toxic, e.g. arsenic, and therefore present challenges in both fabrication and disposal. Second, siliconbased computers are not very energy efficient. They waste a great deal of energy in the form of the heat that they generate and the energy they consume. With these limitations in mind, let’s look at some alternatives to the current computing paradigm.
3 Dept. Of Computer Science & Engineering, SIST
3. Alternatives to Silicon based Computing
Researchers have pursued a number of alternatives to silicon-based computing.These have included biological, optical, and quantum computing. Optical computing is based on replacing with light pulses instead of the electrical signals that carry information through silicon-based computers. Information can travel at the speed of light through information pathways in the optical computing scheme far faster than the commonly used pathways in silicon-based computers. Quantum computers, by contrast, use quantum states of subatomic particles to represent information values. While information is essentially limited to binary (on-off) values in silicon and optical computers, quantum computing permits each information element to carry multiple values simultaneously. Quantum computing also exploits the phenomenon of quantum entanglement, which essentially allows any given information state to exist in two locations simultaneously. This allows the equivalent of instantaneous transfers of information from location to location. It should be noted that a great deal of development works remains to be done before either optical computing or quantum computing can produce practical devices for commercial use. Molecule cascade computing is the newest area in the development of alternatives to traditional computing. This technique is based on forming circuits by creating a precise pattern of carbon monoxide molecules on a copper surface. By nudging a single molecule, it has been possible to cause a cascade of molecules, much like toppling dominoes. Different molecules can represent the 1s and 0s of binary information, making possible calculations. While this technique may make possible circuits hundreds of thousands of times smaller than those used today, it shares with the other alternatives the fact that a number of problems must be solved for it to ever be suitable for practical applications.
4 Dept. Of Computer Science & Engineering, SIST
4. Biological Computing
Biological computing is the use of living organisms or their component parts to perform computing operations or operations associated with computing, e.g. storage. The various forms of biological computing take a different route than those used by quantum or optical computing to overcome the limitations to performance that silicon-based computers face. Rather than focusing on increasing the speed of individual computing operations, biological computing focuses on the use of massive parallelism, or the allocation of tiny portions of a computing task to many different processing elements.Each element in and of itself cannot perform its task quickly, but the fact that there is an incredibly huge number of such elements, each performing a small task, means that the processing operation can be performed far more quickly. Silicon-based computers have used massively parallel processing but will never be capable of the level of massively parallel processing that biological computers can demonstrate. The biological nature of the operation of biological computing also makes it uniquely suited to controlling processes that would require an interface between other biological processes and human technology. The table below compares biological computing with silicon-based computing in several areas, including the component materials, processing scheme, maximum operations per second, presence of toxic components, and energy efficiency. Research in the field of biological computing is focusing on the development of a number of different, though related, forms of technology. While all of these forms share the biological components listed above, they share little else and are best thought of as distant cousins. Some of these technologies will likely be applicable to a variety of problems, while others are best thought of as tools suited to specific purposes.
5 Dept. Of Computer Science & Engineering, SIST
5. DNA Computing
INTRODUCTION TO DNA DNA means DeoxyriboNucleic Acid. The complete set of instructions for making an organism is called its GENOME. It contains the master blueprint for all cellular structures and activities for the lifetime of the cell or organism. Found in every nucleus of a persons many trillions of cells, the human genomes consists of tightly coiled threads of deoxyribonucleic acid (DNA) and associated protein molecules, organized into structures called chromosomes. In humans, as in other higher organisms, a DNA molecule consists of two strands that wrap around each other to resemble a twisted ladder whose sides, made of sugar and phosphate molecules are connected by rings of nitrogen-containing chemicals called bases. Each strand is a linear arrangement of repeating similar units called nucleotides. A nucleotide is composed of one sugar, one phosphate, and a nitrogenous base.
6 Dept. Of Computer Science & Engineering, SIST
7 Dept. Of Computer Science & Engineering, SIST
For users who already have an Internet access and an audio-capable PC. This scenario can take advantage of integration with other Internet services such as World Wide Web, instant messaging, e-mail, etc.
PC to telephone or telephone to PC:
Figure 2 PC to Phone or Phone to PC Scenario In this scenario, PC-callers may reach also the PSTN users. A gateway converting the Internet call into a PSTN call has to be used. Traditional telephone users also can make a call to a PC going through the gateway that connects the IP network with PSTN.
8 Dept. Of Computer Science & Engineering, SIST
Telephone to telephone:
Figure 3 Phone-to-Phone Scenarios The IP network can be a dedicated backbone to connect PSTN. Gateways should connect PSTN to the IP network.
6. BASIC SYSTEM COMPONENTS OF VoIP
There are three major system components to VoIP technology: clients, servers, and gateways.
9 Dept. Of Computer Science & Engineering, SIST
Clients: The client comes in two basic forms. It is either a suite of software running on a user’s PC that allows the user, through a GUI, to set-up and clear voice calls, encode, packetize and transmit outbound voice information from the user’s microphone and receive, decode and play inbound voice information through the user’s speaker or headsets. The other type of client, known as a ‘virtual’ client, does not have a direct user interface, but resides in gateways and provides an interface for users of POTS. Servers: In order for IP Telephony to work and to be viable as a commercial enterprise, a wide range of complex database operations, both realtime and non-real-time, must occur transparently to the user. Such applications include user validation, rating, accounting, billing, revenue collection, revenue distribution, routing (least cost, least latency or other algorithms), management of the overall service, downloading of clients, fulfillment of service, registration of users, directory services, and more.
Gateways: VoIP technology allows voice calls originated and terminated at standard telephones supported by the PSTN to be conveyed over IP networks. VoIP "gateways" provide the bridge between the local PSTN and the IP network for both the originating and terminating sides of a call. To originate a call, the calling party will access the nearest gateway either by a direct connection or by placing a call over the local PSTN and entering the desired destination phone number.
10 Dept. Of Computer Science & Engineering, SIST
The VoIP technology translates the destination telephone number into the data network address (IP address) associated with a corresponding terminating gateway nearest to the destination number. Using the appropriate protocol and packet transmission over the IP network, the terminating gateway will then initiate a call to the destination phone number over the local PSTN to completely establish end-to-end two-way communications. Despite the additional connections required, the overall call set-up time is not significantly longer than with a call fully supported by the PSTN. The gateways must employ a common protocol - for example, the H.323 or SIP or a proprietary protocol - to support standard telephony signaling. The gateways emulate the functions of the PSTN in responding to the telephone's on-hook or off-hook state, receiving or generating DTMF digits and receiving or generating call progress tones. Recognized signals are interpreted and mapped to the appropriate message for relay to the communicating gateway in order to support call set-up, maintenance, billing and call tear down.
7. BENEFITS OF VOIP
11 Dept. Of Computer Science & Engineering, SIST
Voice communications will certainly remain as basic form of interaction among people. A simple replacement of PSTN is hard to implement in short term. The immediate goal for many VoIP service providers is to reproduce existing telephone capabilities at a significantly lower cost and offer a quality of service competitive to PSTN. In general, the benefits of VoIP technology can be the following:
1. Low cost: By avoiding traditional telephony access charges and settlement, a caller can significantly reduce the cost of long distance calls. Although the cost reduction is somewhat related to future regulations, VoIP certainly adds an alternate option to existing PSTN services. Only one physical network is required to deal with both voice/fax and data traffic instead of two physical networks. Having only one physical network has the advantages that lower physical equipment cost, lower maintenance costs. 2. Network efficiency: Packetized voice offers much higher bandwidth efficiency than circuit-switched voice because it does not take up any bandwidth in listening mode or during pauses in a conversation. It is a big saving when we consider a significant part of a conversation is silence. The network efficiency can also be improved by removing the redundancy in certain speech patterns. If we were to use the same 64 Kbps Pulse Code Modulation (PCM) digital-voice encoding method in both technologies, we would see that bandwidth consumption of packetized voice is only a fraction of the consumption of circuit-switched voice. The packetized voice can take advantage of the latest voice-compression algorithms to improve efficiency. 3. Simplification and consolidation:
12 Dept. Of Computer Science & Engineering, SIST
An integrated infrastructure that supports all forms of communication allows more standardization and reduces the total equipment and management cost. The combined infrastructure could support bandwidth optimization and a fault tolerant design. Universal use of the IP protocols for all applications reduces both complexity and more flexibility. Directory services and security services could be more easily shared. 4. Single network infrastructure: When installing VoIP in the office only a single cable is required to the desk, for both telephone and data eliminating separate telephone wiring. 5. VoIP uses "soft" switching: VoIP uses "soft" switching which eliminates most of the legacy PBX equipment. Reducing the cost of installing a communications infra-structure and the maintenance cost once installed. 6. Simple upgrade path: The VoIP PBX technology is software based. It is easier to expand, upgrade and maintain than its traditional telephony counterparts. 7. Bandwidth efficiency: VoIP can compress more voice calls into available bandwidth than legacy telephony. IP Telephony helps to eliminate wasted bandwidth by not transporting the 60% of normal speech which is silence.
IP - the underlying protocol - is supported by most platforms and is independent of the transport protocol used.
8. DEVELOPMENT CHALLENGES
13 Dept. Of Computer Science & Engineering, SIST
The goal of VoIP developers is to add telephone calling capabilities to IPbased networks and interconnect these to traditional public telephone network and to private voice networks maintaining current voice quality standards and preserve the features everyone expects from the telephone. We can summarize the technical challenges as the following. 1. Quality of Service (QoS): The voice quality should be comparable to what is available using the PSTN, even over networks of varying levels of QoS. The following factors decide the VoIP quality: 2. Packet loss: In order to operate a multi-service packet based network at a commercially viable load level, random packet loss is inevitable. This is particularly true with communications over the Internet where traffic profiles are highly unpredictable and the competitive nature of the business drives corporations to load their networks to the maximum. Packetizing voice codec are becoming better at reducing sensitivity to packet loss. The main approaches are smaller packet sizes, interpolation (algorithmic regeneration of lost sound), and a technique where a low-bit-rate sample of each voice packet is appended to the subsequent packet. Through these techniques, and at some cost of bandwidth efficiency, good sound quality can be maintained even in relatively high packet loss scenarios. As techniques for reducing sensitivity to packet loss improve, so a new opportunity for the achievement of even greater efficiencies is presented. This refers to the suppression of the transmission of voice packets whose loss is determined by the encoder to be below a threshold of tolerability at the decoder.
14 Dept. Of Computer Science & Engineering, SIST
This is particularly attractive in the packet based networking world where statistical multiplexing favors the reuse of freed-up bandwidth. 3. Delay: Two problems that result from high end-to-end delay in a voice network is echo and talker overlap. Echo becomes a problem when the round-trip delay is more than 50 milliseconds. Since echo is perceived as a significant quality problem, VoIP systems must address the need for echo control and implement some means of echo cancellation. Talker overlap (the problem of one caller stepping on the other talker’s speech) becomes significant if the one-way delay becomes greater than 250 milliseconds. The end-to-end delay budget is therefore the major constraint and driving requirement for reducing delay through a packet network. Propagation delay (the time taken for the information wave-front to travel a given distance through a given media), jitter buffering, packetization, analog to digital encoding and digital to analog decoding delays are responsible for most of the overall delay. Service and wait time through the switching and transmission elements of the network may be considered trivial given the small packet sizes and relatively wide bandwidths prevalent on the Internet. It is generally true that when considering the achievable quality of a given service, the overall geographic distance traveled by a call is far more important than the complexity of its routing, (i.e. the number of intermediary nodes or "hop-count"). 4. Jitter: Jitter is the variation in inter-packet arrival time as introduced by the variable transmission delay over the network. Removing jitter requires collecting packets and holding them long enough to allow the slowest packets to arrive in time to be played in the correct sequence, which causes additional delay. The jitter buffers add delay, which is used to remove the packet delay variation that each packet is subjected to as it transits the packet network.
15 Dept. Of Computer Science & Engineering, SIST
5. Overhead: Each packet carries a header of various sizes that contains identification and routing information. This information, necessary for the handling of each packet, constitutes ‘overhead’ not present with circuit switching techniques. Small packet size is important with real-time transmissions since packet size contributes directly to delay and the smaller the packet size, the less sensitive a given transmission would be to packet loss. Various new techniques such as header compression are evolving to reduce the packet overhead in IP networks. It is likely that packet based networks, of one form or another, will eventually approach the efficiency, with respect to overhead, of circuit-based networks. 6. User friendly design: The user need not know what technology is being used for the call. He should be able to use the telephone as he does right now. 7. Easy configuration: An easy to use management interface is needed to configure the equipment. A variety of parameters and options such as telephony protocols, compressing algorithm selections, dialing plans, access controls, PSTN fall back features, port arrangement etc. are to be taken care of. 8. Addressing/Directories: Telephone numbers and IP addresses need to be managed in a way that it is transparent to the user. PCs that are used for voice calls may need telephone numbers. IP enabled telephones IP addresses or an access to one via DHCP protocols and Internet directory services will need to be extended to include mappings between the two types of addresses. 9. Security issues:
16 Dept. Of Computer Science & Engineering, SIST
VOIP networks introduce some new risks to carriers and their customers, risks that are not yet fully appreciated. Responding to these threats requires some specific techniques, comprehensive, multi-layer security policies, and firewalls that can handle the special latency and performance requirements of VoIP. It is important to remember that a VoIP network is an IP network. Any VoIP device is an IP device, and it's therefore vulnerable to the same types of attacks as any other IP device. In addition, a VoIP network will almost always have non-VoIP devices attached to it and be connected to other mission-critical networks. Every IP network, regardless of how private it is, eventually winds up connected to the global Internet. Even if it is not possible to directly route a packet from the "private" network onto the Internet, it is extremely likely that some host on the "private" network will also be connected to a less private network. Compromising this host provides an attacker with a gateway into the presumed secure private network. It's important, therefore, to secure all IP networks, but VoIP networks have special security requirements. Specific techniques, comprehensive policies, and VoIP-capable firewalls are needed to do the job right.
9. VOIP APPLICATIONS
17 Dept. Of Computer Science & Engineering, SIST
Cross-platform connections Some VoIP products, including Skype and Gizmo Project, run on Windows and Linux, while Apple’s iChat AV runs only on OS X. Skype, SightSpeed, Gizmo Project, and iChat AV allow you to host either multiparty voice or videoconference calls. Unlike expensive high-end conferencing systems designed for large businesses, which are often connected to a telephone system, these simple desktop VoIP apps can make conferencing easier —and more affordable. All of these applications allow you to call other Internet users for free. But if you want to call somebody using his or her telephone number, as permitted by Skype, Gizmo Project, and the Wengo plug-in, you’ll pay a basic, per-minute fee. At this writing, neither iChat AV nor Sigh Speed permits computer-to-phone calling. Advanced features cost money While you can make basic calls for free, more-advanced features will cost you. For instance, Skype’s voice-mail feature carries a small monthly charge. Obtaining a permanent phone number from Skype (called a SkypeIn number) involves an additional fee. Also for a fee, Gizmo Project allows you to forward your incoming calls to another telephone, such as your cell phone, and SightSpeed offers extended conferencing and video-messaging features for paid subscribers. iChat AV users can’t call traditional phone numbers, but they can call each other, using securely encrypted audio channels on the Internet if all participants are .Mac subscribers. Once you become accustomed to a desktop VoIP tool, you may find that VoIP calling becomes a part of your daily routine. After all, it’s a lot easier to dial a Skype buddy by double-clicking on a name than it is to look up a number in Address Book and manually punch it in on your telephone’s keypad. If you’re into multiplayer Internet games, using a tool like Skype to keep in touch with your teammates is nice, as it relieves you from having to type text-chat
18 Dept. Of Computer Science & Engineering, SIST
messages during the game. And if you have relatives in other countries, talking to them over the Internet will cost you a lot less than placing international longdistance calls.
VoIP in Military Applications Military organizations located worldwide are currently transitioning their telephony infrastructure from legacy TDM to Next Generation Networks (NGN) based on VoIP technology. The reasons for the migration taking place within the military have many similarities to those of the migration in the commercial telephony space. Some of these similarities include lower OPEX resulting from having a single consolidated network for data and telephony, and the ability to deliver new services quickly. However, deployments within the military can have additional specific benefits when moving to VoIP. Voice over IP, as its name implies, traverses over IP networks which, when designed correctly, can be more resilient than TDM networks. IP networks are easier to deploy and manage compared to their older TDM counterparts. In the past, significant investments were made in legacy telephony equipment by military organizations. In most cases, this investment included military purpose built TDM equipment. As a result, the process of migration to NGN will be gradual. It may take many years before full end-to-end VoIP communication can be realized - where all handsets are SIP based and TDM trunks are eliminated.
Mobile VoIP Today, as cost efficient communications increase in demand, Mobile VoIP technologies that turn a mobile device into a SIP client and use a
19 Dept. Of Computer Science & Engineering, SIST
data network for sending and receiving communications, have become increasingly important to users. The mobile VoIP market is expected to be worth $32.2 billion by 2013 and by 2019, half of all mobile calls will be made over allIP networks, according to recent industry reports. Mobile VoIP provider REVE Systems offers operators support for the shift to mobile VoIP with their iTel Mobile Dialer Express, a mobile application that makes it possible to use VoIP via any mobile phone and that can be branded by operators. iTel Mobile Dialer Express supports GPRS, Wi-Fi and Bluetooth for Internet connectivity and can run on any phone on Symbian or Windows Mobile 5 and 6 platforms.
10. IS VOIP THE FUTURE OF TELECOMMUNICATIONS?
20 Dept. Of Computer Science & Engineering, SIST
VoIP means that the technology used to send data over the Internet is now being used to transmit voice as well. The technology is known as packet switching. Instead of establishing a dedicated connection between two devices (computers, telephones, etc.) and sending the message "in one piece", this technology divides the message into smaller fragments, called 'packets'. These packets are transmitted separately over a decentralized network and when they reach the final destination, they're reassembled into the original message. VoIP allows a much higher volume of telecommunications traffic to flow at much higher speeds than traditional circuits do, and at a significantly lower cost. VoIP networks are significantly less capital intensive to construct and much less expensive to maintain and upgrade than legacy networks (traditional circuit-switched networks). Since VoIP networks are based on Internet protocol, they can seamlessly and cost-effectively interface with the high technology, productivity-enhancing services shaping today's business landscape. These networks can seamlessly interface with web-based services such as virtual portals, interactive voice response (IVR), and unified messaging packages, integrating data, fax, voice, and video into one communications platform that can interconnect with the existing telecommunications infrastructure. Industry experts see VoIP as a tool that will become the standard platform for the international calling market. It is strongly believed that the profit realization VoIP will trigger in the global telecommunications industry will dwarf the impact of the now ancient "digital revolution". As with any promising new technology, a myriad of companies are trying to climb aboard the VoIP bandwagon. Currently, however, the industry is characterized by a high degree of confusion. Most companies, including large resource-rich national and international telecommunications carriers, are experiencing enormous difficulty in building effective international VoIP
21 Dept. Of Computer Science & Engineering, SIST
networks. They are unable to harness the power of VoIP or effectively communicate the benefits of VoIP to their customers. We mustn’t ignore the problem of scalability. The system has to be designed so that it can grow. And each segment within the system must be able to grow. Creating an architecture that can handle billions of minutes of use per month requires a solution with high call processing capabilities. If we are looking at a global solution, we have to start from the beginning with a global approach. And that’s one of the reasons why a fully implemented solution won’t be available tomorrow.
11. CONLUSION
22 Dept. Of Computer Science & Engineering, SIST
Data traffic has traditionally been forced to fit onto the voice network (using modems, for example). The Internet has created an opportunity to reverse this integration strategy – voice and facsimile can now be carried over IP networks, with the integration of video and other multimedia applications close behind. The Internet and its underlying TCP/IP protocol suite have become the driving force for new technologies, with the unique challenges of real-time voice being the latest in a series of developments. Telephony over the Internet cannot make compromise in voice quality, reliability, scalability, and manageability. Future extensions will include innovative new solutions including conference bridging, voice/data synchronization, combined real-time and message-based services, text-to- speech conversion and voice response systems. The market for VoIP products is established and is beginning its rapid growth phase. Producers in this market must look for ways to improve their time-to market if they wish to be market leaders. Buying and integrating predefined and pre-tested software (instead of custom building everything) is one of the options. Significant benefits of the “buy vs. build “ approach include reduced development time, simplified product integration, lower costs, offloading of standard compliance issues, and fewer risks. Software that is known to conform to standards, has built-in accommodation for difference in national telephone systems, has already been optimized for performance and reliability, and has “plug and play” capabilities can eliminate many very time-consuming development tasks.
12. REFERENCES
23 Dept. Of Computer Science & Engineering, SIST
1. 2. 3. 4. 5. 6.
Computer Networks by Andrew S.Tanenbaum Internetworking with TCP/IP by Douglas E.comer www.iec.org.com www.telogy.com www.rad.com www.mailto:[email protected]
24 Dept. Of Computer Science & Engineering, SIST
