Quality of Service - Case Study Example

Introduction Quality is a difficult term to define exactly as it lacks definite definition for it. For its clear definition, it is very much dependent upon where it is being used, how it is being and importantly, why it is being used. In Computer Networking field, Quality of Service refers to control mechanism for resource reservation. It is about different priority to different application, users or data flows so that a certain level of performance to a particular data flow (Voice, Video etc) could be defined. It is important and very helpful in circumstances where a required bit rate, delay or packet dropping probability needs to be guaranteed. Quality of Service is helpful when network capacity is insufficient as compared to network users. For same reason, QoS is more in demand on Corporate LANs, private networks, Organizations Intranets etc. A network with varied types of data travelling around, like students playing games over network may hinder traffic on some other important types of data. This is usually the motivation to ensure favorable traffic over less important data traffic. QoS is relatively less in demand on the internet and ISP Networks. Generally, how quickly we receive text; audio or video depends upon the bulk of the media. However, it’s very interesting to note that if the ISP provides QoS that favors voice then voice quality would be great and other media types may suffer in comparison. Conventional internet routers and LAN switches lacked ability to provide QoS. This simplicity in equipment made it faster and relatively cheaper as well but sophistication of QoS was required. Through the years, QoS mechanisms have become more and more sophisticated and these mechanisms are available from small LANs up to giant networks. It’s important to consider elements of QoS in order to understand its impacts on video and voice over IP. Bandwidth: Bandwidth refers to the data rate supported by a network connection. The most common measurement we do for bandwidth is bits per second (bps) i.e. effective number of data units transported per unit time. 56 Kb/s modem were quite famous and extensively used for long time before bandwidth become cheaper and easily available. Similarly, the rated bandwidth for DSL is 1.4 Mbps down and 400 Kbps up. Cable modems are typically rated at 1.5 to 3 Mbps down and 400 to 600 Kbps up. It’s very obvious that Multimedia applications (like voice and video) usually require high bandwidth as compared to other general applications. For example, MPEG2 video requires around 4Mbps, while MPEG1 video requires about 1-2 Mbps. Network technologies that do not support such high bandwidth cannot play multimedia contents. For instance, Bluetooth technology version 1 only supports a maximum bandwidth of 746 kbps and that means that devices relying on Bluetooth technology version 1 for connectivity cannot play MPEG1 videos. We can say that better bandwidth is directly proportional to better QoS for each device. With more and more multinationals growing their businesses across the globe and also strong trend about outsourcing work to cost-effective areas, there is a boom in the use of conferencing applications like videoconferencing webinars/ webcasts. In current circumstances, there will be delays and loss but many problems can be linked to how bandwidth is handled by network operators and businessmen. Modern multimedia applications have grown and become sophisticated that they can nearly make use of unbound amount of network bandwidth. Allocation of more bandwidth to the applications can produce improvement in latency, fidelity, resolution, reliability and other such important service features. However, regardless of the bandwidth available, there is almost always a potential value to having more devoted to the application (Yamaki, et al,. 1996). Since the available bandwidth may be distributed across the possible applications and used in a variety of ways so there will be tradeoffs in allocations. The problem of QoS allocation in a distributed environment has always been an area of focus in work on multimedia systems and networks (Nahrsted, Vogel 1995). With more and more focus shifting towards mobility in computing, wireless networks have become a favorite ground for researchers. Adaptive Multimedia is promising in wireless/mobile networks since it mitigates fluctuation of resources caused by mobility in these networks. However, bandwidth adaptation entails the bandwidth degradation for some application (Xiao et al 2002). Delay: Delay, in simple terms, can be defined as the time interval elapsed between the departures of data from the source to its arrival at the destination. Or in terms of communication system, it refers to the time lag between the departure of signal from the source and its arrival at the destination. This can range from a few nanoseconds or microseconds in local area networks (LANs) to about 0.25 s in satellite communications systems. Greater delays can occur as a result of the time required for packets to make their way through land-based cables and nodes of the Internet. Because of the clock synchronization problem, it is difficult to measure one-way delays; therefore, round-trip (i.e. forward and return paths on the Internet) delays are used. Delay/Latency is very well explained by Bhatia, et al (2006) as the amount of time it takes for speech to exit speaker’s mouth and reach the listener’s ear. As we know that light travels at a certain speed in vacuum to reach us. Similarly electricity that reaches our homes and offices is dependent upon speed at which electrons travel. In same way, propagation delay is the amount of time it takes for the signal to travel from sender to receiver. In order to better utilize the network resources in wireless networks to locate mobile users, new paging schemes are being presented. Paging systems provide reliable information transfer that reaches many places that are not reachable by other types of mobile or satellite systems. As with each day, demand of wireless services grows rapidly, the signaling traffic caused by location update and paging increases accordingly, which consumes available resources. Delays and costs are two key issues in the paging issue. Of the two factors, paging delay is very important as the Quality of Service (QoS) requirement for multimedia services (Wang et.al, 2001). Jitter: Jitter is the term used for variation in time for packets receiving at destination. Jitter could due to network congestion, router change or timing drift. Depending upon the type of multimedia application, jitter may or may not be significant. Audio and video conference applications are not tolerable to jitter due to very limited buffering time in live presentation. We can have a real life example of buffering from our daily usage of internet, when we click on the video it is partially loaded before we access it. This is known as "buffering". This little delay helps in loading and playing the video instantaneously. The challenges however, are two-fold in buffering. First, both buffer underflow and buffer overflow (buffer underflow is a state when buffer used to communicate between two devices is fed with data at a lower speed than the speed data is being read from it whereas a condition where a portion of memory being used as a buffer has a fixed size but is filled with more than that amount of data) may occur and lead to jittering. Hence, a suitable buffer size must be determined. Second, as we are interested in providing continuous video streaming service, the initial playout delay should be minimized. More and more research is going on for Jitter control in networks. Different researches with aim at delay Jitter, where aim is to minimize the differences between delay times of different packets, and rate Jitter, where the aim is to minimize the differences between inter-arrival times. Mansour, Patt-Shamir (2001) studied Jitter control in networks with guaranteed Quality of Service (QoS) from the competitive analysis point of view: they proposed on-line algorithms that control jitter and compare their performance to the best possible for any given arrival sequence. Loss: As it is quite obvious from the name, loss means the packets that didn't make it to the destination in a specified time period. Different methods can be used to reduce chance of loss, either a separate dedicated channel could be used or technique to re-transmit the data from loss recovery. As with other areas of research for network robustness, loss-tolerant QoS have been heavily researched as well. For real time-applications, the QoS important metrics are delay and loss. To ensure short delay and loss-free guarantee, the peak-rate bandwidth is commonly used especially for variable bit-rate flows whose carried burst size is quite large (Koubaa, Song 2004). Reliability: Applications with Real-time processing make packet retransmission impossible. Most multimedia applications are tolerant to error up to a certain extent. Successful delivery of all the multimedia data packets is important. Although system reliability depends upon many factors in the network, it’s easy to understand that system reliability increase when failure rate decreases. Frame size and Frame rate: Frame size defines the size of video image on the display screen. Bigger frame size requires greater bandwidth. While adapting to QoS, frame size can be modified according to available QoS support, it means that at network congestion time, frame size could be reduced for video transmission. Frame rate means the number of frames sent to network per unit time. Higher frame rate is achieved at higher bandwidth. Frame rate can also be modified to lessen the bandwidth requirement but it will reduce the video quality. Image Clarity: Image clarity is the user perception about quality of image for a certain delivery of multimedia content. High frequency components of an image can be suppressed without any noticeable loss in image quality because human eyes are less sensitive to high frequency components of an image. Reducing high frequency components of an image technique is used when there is a need to blur an image or an image can be sharpened by increasing the magnitude of its high frequency components. Figure below is taken from Frequency domain methods and it shows ideal low pass filter that would retain all the low frequency components and eliminate all the high frequency components. Audio Quality: Audio clarity is linked with sampling rate per second. A higher sampling rate would mean a better audio quality. For Example, audio encoded in 128kbps is higher in quality than the one encoded in 16kbps. How does QoS work? A Short Description QoS works by prioritizing packets, sometimes slowing unimportant packets down, or in the cases of extreme network traffic, throwing them away entirely.  This leaves room for important packets to reach their destination as quickly as possible.  Basically, once the router is aware of data capacity, it can shape traffic accordingly by prioritizing important packets first, then using any leftover space to fill the pipe up in descending order of importance. It is important to note that QoS cannot speed up a packet, basically takes up total available bandwidth, calculate how much of the highest priority data it has and then prioritizes it accordingly. QoS packets may be prioritized by a number of criteria, prioritization could be done by application but the most common techniques is the prioritization by the consumer himself. Morris (2004) explains it in general terms that just like priority queues in supermarkets can produce happy customers, the same analogy obtains for network devices. Since QoS prioritizes different data streams over a network so it’s important to consider the issues that may occur in video or VOIP implementation. The issue mainly is how to guarantee that packet traffic for a voice of video is not delayed or dropped due to interference from other traffic on the network. End to end delay is present in all types of communication but its main effect is on voice over IP because it causes discomfort and conversational difficulty. Delays below 100mS are acceptable whereas delays above that would create increasingly more communication problem. Voice, video, and other time-sensitive technologies onto traditional data networks requires that the streams be delivered on time with minimal loss and consistent delay. A delay or loss of packet in a voice stream, for example, causes a break or chop in the conversation. Since a voice of a video packet data is only relevant at one point in conversion so it doesn’t need to be retransmitted if it gets lost. Traditional data networks are less concerned with on-time delivery and loss. Since the usage of Audio and Video data is diverse so following factors are very important to be considered in QoS for video and VoIP. Confidentiality and Integrity: Confidentiality and integrity for audio and video data is very important with respect to their particular context. It means the security and access rights restrictions for audio and video streams. Content type defines whether this feature is useful or not. Security is very vital in case of audio/video conference sharing confidential information; on the other hand security is not very important for general information like internet broadcast. The funny thing about QoS is that it tends to be known only after the fact; you use some application and you experience the QoS as you use it. Somehow in similar terms, same principle is applicable to QoS, poor QoS shows itself in excessive delay and possibly lost packets. In a more general sense, Main features that can be set as benchmarks for QoS usefulness are mentioned below: Throughput: Amount of available bandwidth between two network points of interest would determine the bandwidth available for use. Availability: Fraction of time during which service is available between two points, it would allow to judge the time during which service could be used. Delay: Length of time taken to transmit data from one point to another, it is just like time from the speaker mouth to listener’s ear. Delay variation: Difference in the delay for two packets, it means how much time delay there is in arrival time of two packets. Loss: Ratio of packets received to packets sent, it would determine the loss to success ratio out of total packets sent and received. These simple elements are sufficient to put together very complex service level agreements (SLAs). Some service providers now supply stringent SLAs with proactive tariff rebates—all part of the increased importance of service differentiation. As time goes on, we can expect to see this enlarged role of SLAs emerge increasingly within enterprise networks. During the recent Internet stock bubble, articles in the trade press frequently said that, in the near future, telephone traffic would be just another application running over the internet. Such statements gloss over many engineering details that preclude voice from being just another Internet application. IP networks are designed to support non-real time applications, such as file transfer or e-mail. These applications are characterized by their bursty traffic and high Bandwidth demands at burst times, but they are not very sensitive to delay or delay variation (Chandra 2004). REFERENCES Morris, B. Stephen, 2004. Quality of Service, Part 2 of 2: Managing Enterprise QoS [Online] Available at: http://www.informit.com/articles/article.aspx?p=344237 [Accessed 7 April 2010] Bhatia, M. Davidson, J. Kalidindi, S. Mukherjee, S. Peters, J. VoIP: An In-depth Analysis [Online] Available at: http://www.ciscopress.com/articles/article.asp?p=606583 [Accessed 7 April 2010] Yamaki, Hirofumi. Wellman P., Michael. Ishida, Toru. A Market-Based approach to Allocatin QoS for Multimedia Applications In: Second International Conference on Multiagent Systems, 1996 Nahrstedt K,. Smith, M, J. The QoS Broker, IEEE Multimedia 2(1), pp.53-67 Wang, Wenye. Akyildiz, F, Ian. STÜBER, L, Gordon. Chung, Boo-Young, Effective Paging Schemes with Delay Bounds as QoS Constraints inWireless Systems, Wireless Networks 7, pp.455-466, Netherlands Xiao Y., Chen C.L.P., Wang B. Bandwidth degradation QoS provisioning for adaptive multimedia in wireless/mobile networks, Computer Communications, 25 (13), pp. 1153-1161. (2002) Mansour, Yishay. Patt-Shamir, Boaz. Jitter control in QoS networks, IEEE Press 9(4), pp.495-502 (2001) Koubaa, Anis. Song, Ye-Qiong, Loss-Tolerant QoS using Firm Constraints in Guaranteed Rate Networks, IEEE Real-Time and Embedded Technology and Application Symposium (2004) Chandra, Sarat, 2004 QoS in VoIP. Available at: http://www.it.iitb.ac.in/~sarat/seminar/seminar_QoS_VoIP.pdf [Accessed 8 April 2010] Owen, Robyns. 1997, Frequency Domain Methods. Available at: http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/OWENS/LECT5/node4.html Read More