Capacity Evaluation of Multi-User Multi-Cell WIFI - Coursework Example

Capa Evaluation of Multi-user multi-cell WIFI In this paper i present systematic experimental study on evaluation of the capacity of wireless networks i.e. IEEE 802.11 by investigating the multiple access schemes. We analyze the access points’ density and capacity in different propagation conditions i.e. indoor and outdoor. We also analyze the effects when access points transmission power increase or decrease for capacity and co-channel interference form closer neighboring access points. The study will also be inclusive of investigation of the relationship between total cell throughput and mean user throughput. A multiple access scheme is a reminiscent of the conventional multiple-user networks but however in this type of access, scheme cells rather than users compete for access. The capacity of the network in this case is analyzed and an optimization method where probability distribution for access is binary will be appropriate. Categories and Subject Descriptors: Performance of systems and attributes, Measurement techniques, capacity evaluation General terms: Performance, Evaluation and measurement Keywords: multiple access schemes, Intercell interference, throughput, access points, Wi-Fi Introduction In recent years there has been an explosion in the availability and use of wireless internet access both for PCs and mobile devices that are Wi-Fi enabled. However a problem for Wi-Fi enabled devices users, service providers and application designers is the seeking out and supporting the connectivity option that provides the best and most reliable performance. Wi-Fi is a wireless technology that uses radio waves to provide a wireless high-speed internet and network connections. It is based on the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards. This standard has no physical wired connection between sender and receiver and uses the frequency within the electromagnetic spectrum that is associated with radio wave propagation. An electromagnetic field is created when a radio frequency is applied to an antenna which able to propagate through air. An access point is the cornerstone of any wireless network and its primary job is to broad cast a wireless signal that computers can detect and connect to. Today’s wireless networks are optimized from the point of view of the physical layer but are however hindered by inefficient reuse of the spectral source over different cells. Therefore spectral re-use schemes will play an important role in mitigating channel interference i.e. multiple access schemes. These schemes enable several multiple users to gain access to the network and use it simultaneously. Such a scheme should be able to handle several users without interference, maximize the spectral efficiency and provide smooth handover between cells. A wireless access points is abridge that serves to join wireless devices and clients to a wired Ethernet network. Each access point supports up to a maximum number of users without loss of connectivity. This is referred to its capacity. Also the access points have the ability to serve a specified area radius depending on the specifications of the hardware i.e. access points’ density. Access points can be set up in specific areas depending on the purpose and the users. For instance, it could be indoor if the users will be stationed indoor most of the time or outdoor unit if it will be available to users in a surrounding environment. However, outside users can still access the indoor access points and vice versa but the density will be affected and hence the capacity. IEEE Standards (Summary) Depending on the propagation environment, Wi-Fi network can be deployed in either high-density of low density. High density design refers to any environment where the client services are positions in densities greater than the coverage expectations of a normal network. This is a typical indoor office environment with signal characteristics for signal attenuation. User density is the critical factor and the bandwidth achieved per radio cell, the capacity and their connection characteristics for instance, band signal, duty cycle speed, radio type and SNR that occupy that cell determines the overall bandwidth available per user. Spectrum available for wireless networks For IEEE 2.4 GHz 802.11b,it uses DSSS Modulation, requires 22MHz band for a single network and can support up to 3 non-overlapping simultaneously Wi-Fi Networks with an 83.5MHz. Figure 1: Available spectrum for IEEE 2.4GHz 802.11b For IEEE 5GHz 802.11a; it utilizes OFDM Modulation, requires only 16.6 MHz band for single network and supports up to 12 non-overlapping networks simultaneously with a spectrum of 300MHZ Figure 2: Spectrum for IEEE 5GHz 802.11a The signal range for any wireless system is governed by the following: • RF power transmit level • Signal Energy that is required to recover the transmitted symbol • Environment The difference between communications operating at 2.4 and 5GHz is the communication range achieved between the AP and the connecting device. With the variables stated above being held constant, 2.4 GHz frequency offer roughly double the range of those operating in the 5GHz band. Analyzing the performance using measured data We assume measured performance data collected in a typical office environment 265 foot by 115 foot rectangular with conference rooms closed offices and walls as well as semi-open cubicle spaces. For the 802.11a system data was sent between two PC cards (TX power=15dBM, =32mw) one as fixed AP and the other as a mobile station with output power of 14dBm. For the 802.11b system comprised as AP and a PC card with output power of 15dBm Data Link Rate results (for test environment with one AP and a mobile Station) Figure 3: Data Link Rate results with one AP and a mobile Station Throughput Results (for test environment with one AP and a mobile station) Figure 4: Throughput Results with one AP and a mobile station In such an indoor environment, the access point is placed in an area that has high user sensitivity e.g. conference rooms while other common areas are left with less coverage. These high density areas appear in clusters so as to achieve maximum capacity of the area i.e., the higher the number of usage the higher the density per square feet. For example, a room with occupancy of 32 users will require a user density of on user per 28 square feet. In an indoor environment such as an auditorium, densities increase dramatically. Therefore, all the dimensions of space are crucial in getting the free path loss of the signal form the access point. However the densities are unevenly distributed over the entire indoor space and hence the APs will be exposed with an excellent view of the room and the entire user device will be closely packed surrounded with attenuating bodies. However, for indoor environment the signal does not predictably lose energy so is difficult to calculate Path Loss. For an indoor environment, the power of the signal at the receiver is related to that at the transmitter as shown below. Where C=speed of light, f= frequency, N= coefficient for path loss. In any wireless network, radio resource management is very essential. This is done by the radio resource management software that monitors the total bandwidth used for transmitting and receiving traffic , amount of traffic coming from other adjacent sources Amount of non-802.11 that is interfering with the current signal Strength of the incoming signal and signal to noise ratio for all the clients connected Number of adjacent access points Therefore for an 802.11 network, one of the important aspects that are monitored for network best efficiency is the transmission power control. In this context we refer a maximum TX power = 20 dBm (=100 mw). These access point transmission power is controlled dynamically on a real-time basis. This is done through an algorithm that increases and decreases the power depending on the changes on the RF environment. This power control seeks to lower the power to reduce interference but in a case the access point becomes disabled, the power of the surrounding access points can be increased. Power Considerations The limitation in range is caused by the more severe path loss of the 5GHz spectrum. However by increasing the power of a 5GHz system approximately 4 times, similar ranges can be achieved by 2.4GHz systems. 802.11a system (5GHz) is more power efficient over small areas while 802.11b (2.4GHz) is more efficient over greater distances. Therefore when the transmission power of the adjacent access points is increased, the network Wi-Fi capacity increases since the signal strength increases and vice versa. This is a phenomenon where the transmission from on access point bleeds into other access points’ ranges that are on the same channel. This causes interference and reduces the available spectrum and the performance of those devices. This co-channel interference can so cause collisions in the transmission and hence corrupting the frames in transit. The figure below best illustrated how access points in the same channel interfere with each other. Figure 5: Basic Co-channel interference Therefore an increase in the access points’ transmission power results to an increase in the co-channel interference. An increase in the transmission power will result to an increase in the access point density. And in a high density network, propagation will be good but in this case, co-channel inference will be more likely. Therefore the effects of this kind of interference in a Wi-Fi network can be avoided by using non-overlapping channels and an aspect of natural attenuation in order to isolate individual cells. System Capacity under co-channel inference using measured data This analysis will be based on an 8-cell system. System Capacity in this context refers to the throughput of an entire WLAN comprised of many cells. Co-Channel interference is less pronounced in for 802.11a systems as compared to 802.11b because of more channels. The mean Cell Throughput under CCI using measured data Data from single AP-User entered into a system capacity model to evaluate the system capacity of the 8-cell WLAN system above Figure 6: The mean Cell Throughput under CCI using measured data The average cell throughput varies differently as shown above for both wireless standards i.e. 802.11b and 802.11a Capacity of a System under CCI using the measured data Capacity of a system as the mean cell throughput (shown above) multiplied by the number of cells Figure 7: System Capacity under CCI using measured data In any communication system, throughput refers to the average rate of messages delivered successfully over a channel. It’s measured in bits per second or data packets per second. Therefore total cell throughput refers to the average rate of messages that are sent successfully over the network while mean user throughput is the average rate of message user are sending form their devices across the wireless network. Figure 8: Throughput Overview Throughput is a fundamental metric when discussing data communications. It depends on the types of data traffic and supporting communication system. Considering two antennae system with many low data users and one high data 153.6kbps data user, one could expect an increase in total system data throughput of approximately 2% while for a 4 antennae system with many low data users and one high data of 307.2kbps data user could expect an increase in total system data throughput of around 5.0%. Therefore in a wireless network, total cell throughput relates directly with the mean user throughput but this mean user throughput can be affected by loss of packets due to network congestion, bit errors or unfair scheduling algorithms that favors some users to others. Figures 9 show the mean user throughput versus users and the average throughput versus the access point density. It’s very clear that the average user throughput varies inversely with the number of users in that the access point. As the number of users in an access point increases, the density of that point will decrease and hence the signal strength. Therefore the number of successful messages sent from the users devices over that network will decrease. This explains the relationship displayed in the graph. However the average throughput density varies directly with the density of the access points. An increase in the density of an access point will result with an increase in signal strength. This mean more messages will be sent successfully over the network hence an increase in average throughput. Figure 9: Mean User Tput vs. UEs On figure 10, the effective system throughput varies inversely with the distance from the access point. Each access point has strength of broadcasting up to a certain range. Therefore, as one moves away from the access point, the signal strength will always become weaker since the waves gets attenuated due to path loss. Therefore the number of messages sent over that channel will be limited and hence this will explain why the effective system throughput varies inversely with the access point density. Figure 10: Effective system throughput vs. range SINR refers to Signal to Interference plus Noise Ratio. It is commonly used in wireless communication to calculate the quality of the connections. Normally, strength of a signal fades with an increase in distance. This is due to the fact that the wave attenuates as it propagates away from the access point i.e. path loss. The attenuation is due to the following, refraction, reflections, diffraction, free space loss and absorption. It could also be influenced by terrain, environment, propagation medium, distance between the transmitter and receiver and height and location of antennas. SINR can be calculated from Where P = represents the received power). I = represents the interference power of other simultaneous transmissions N = represents the background noise power Path loss can be expressed using path loss exponent in the case of wireless communications. The value of this exponent ranges from 2-4 depending whether the propagation is in free space or in lossy environments. However in some environments like indoors and buildings, this exponent can reach values ranging from 4-6. The following formula represents path loss in decibels: Where L=path loss in decibels n= path loss exponent d= distance between access point and accessing device C= constant for system losses Consider a Wi-Fi network with an AP using CSMA/CA MAC protocol and assume a traffic downlink using a single user per AP 2. The coverage area of the AP will be thus be defined as the set of user locations for which SINR is greater than the minimum SINR threshold value. The corresponding maximum coverage range of AP will be the mean distance from the AP to users located on the contour for which SINR= bit rate. Therefore for the Co-channel Aps, a pair will be considered to be the common contention domain if they receive each other’s signal with a power greater than the threshold for carrier-sensing. According to the CSMA/CA MAC protocol, one AP will not transmit if another AP in its contention domain is also transmitting at the same time. This will therefore cause interference. Therefore, the mean SINR over AP coverage will take the following relationship: Where; Pu=transmission power of the second AP Lu= average path loss on the link between the transmitter and receiver Nu= noise power at the receiver In a wireless network, if the SINR perceived at the receiver exceeds the minimum SINR threshold of AP, then transmission is considered successful. However the data rate that can sustain a transmission depends on the SINR value. Multiple data rates are available due to tremendous advances in in wireless technologies. These are four data rates for 802.11b (1, 2, 5.5, 11Mbps) and eight data rates for 802.11a/g (6, 9, 12, 18, 24, 36, 48, 54 Mbps). It is important to note that the higher the SINR, then the higher the data rates at which that transmission can sustain. Therefore for a give SINR, one will prefer to choose the highest possible data rate the transmission could sustain i.e. on that allows the given SINR to be decoded correctly. This is in order to maximize the throughput of the system. The value of the SINR plays a vital role in determining whether or not a transmission will be successful and /or the data rate the transmission can sustain. This therefore will depend on the value of SINR chosen since the higher the value of minimum SINR threshold, the higher the spatial reuse but cumulative interference by concurrent transmission outside may corrupt the transmission from sustaining a higher data rate. Conclusion A wireless access points is abridge that serves to join wireless devices and clients to a wired Ethernet network. According to IEEE standards, there exist three standards for wireless networks: 802.11a, 802.11b and 802.11. 802.11b requires 22MHz band for a single network and can support up to 3 non-overlapping simultaneously Wi-Fi Networks with an 83.5MHz while 802.11a requires only 16.6 MHz band for single network and supports up to 12 non-overlapping networks simultaneously with a spectrum of 300MHZ. Analyzing the evaluation performance of a wireless network and assuming an indoor office environment, it is found that the data link rate for 802.11a is higher than 802.11b for the same range. Therefore in such an indoor environment, the access point needs to be placed in an area that has higher user sensitivity. However, for indoor environment the signal does not predictably lose energy so is difficult to calculate Path Loss. An 802.11a system (5GHz) is more power efficient over small areas while 802.11b (2.4GHz) is more efficient over greater distances. Therefore when the transmission power of the adjacent access points is increased, the network Wi-Fi capacity increases since the signal strength increases and vice versa. Co-channel interference refers to a phenomenon where the transmission from on access point bleeds into other access points devices range that are on the same channel. This causes interference and reduces the available spectrum and the performance of those devices. Therefore an increase in the access points transmission power results to an increase in the co-channel interference and hence access point density. However this can be avoided by using non-overlapping channels in order to isolate individual cells. Throughput is a fundamental metric when discussing any data communications and it depends on the types of data traffic and supporting communication system. In a wireless network, total cell throughput relates directly with the mean user throughput but this mean user throughput can be affected by loss of packets due to network congestion, bit errors or unfair scheduling algorithms that favors some users to others. Average throughput density varies directly with the density of the access points and an increase in the density of an access point will result with an increase in signal strength. This means more messages will be sent successfully over the network hence an increase in average throughput. SINR refers to Signal to Interference plus Noise Ratio and is commonly used in wireless communication to calculate the quality of the connections. Normally a wave attenuates as it propagates away from the access point i.e. path loss. This is due to refraction. Reflections, diffraction, free space loss and absorption. The higher the SINR, then the higher the data rates at which that transmission can sustain. References Ababneh, I, Al-Ghadi & Mardini, W, 2012 ‘Performance study of SINR scheme for Vertical Handoff in wireless networks’. Paper presented at the Computer Science Department, Jordan University of Science and Technology Irbid, Jordan. Bonald, T, Ibrahim & Roberts, J, 2009 ‘Traffic Capacity of Multi-Cell WLANs’. Paper presented at the Orange Labs, Orange, France. Burke, J & Zeidler, J. 2010 ‘Data Throughput in a Multi-Rate CDMA Reverse Link: Comparing Optimal Spatial Combining vs. Maximal Ratio Combining’. Paper presented at the Department of Electrical and Computer Engineering, University of California, San Diego. Carrier-class 2.4/5GHz 802.11n Outdoor Access Point with integrated small cell, 2013. Available from Castenada, Mezghani & Nossek, 2011 ‘On maximizing the sum network miso broadcast capacity’. Paper presented at the Munich University of Technology, Munich. Cisco High Density Wireless LAN Design Guide 2013. Available from < www.cisco.com/en/US/prod/collateral/wireless/ps5678/ps10981/design_guide_c07-693245.html> Cisco Wireless LAN Controller Configuration Guide 2013. Available from < http://www.cisco.com/en/US/docs/wireless/controller/7.2/configuration/guide/cg_rrm.html> Cruz, R & Ehsan, N 2008 ‘On the Optimal SINR in Random Access Networks with Spatial Reuse’ . Paper presented at the Department of Electrical and Computer Engineering, University of California, San Diego. Ha, T & Vu, 2013, ‘Channel capacity of multi user TR-MIMO-UWB communications system ’. Paper presented at Computing, Management and Telecommunications (ComManTel) International Conference, IEEE, Ho Chi Minh City, Vietnam. Hamad, Hasan & Kawser, 2012, ‘Downlink SNR to CQI Mapping for Different Multiple Antenna Techniques in LTE’, International Journal of Information and Electronics Engineering, Vol. 2, No. 5, pp. 1-4. Hou, J & Lin, T, 2010 ‘Interplay of Spatial Reuse and SINR-determined Data Rates in CSMA/CA-based, Multi-hop, Multi-rate Wireless Networks’. Paper presented at the Department of Computer Science, University of Illinois, Illinois. How much capacity does a wireless n access point have? 2011. Available from < http://www.securedgenetworks.com/secure-edge-networks-blog/bid/54090/How-much-capacity-does-a-wireless-n-access-point-have> Joung, J., Jang, Y & Kim, E., & S, W, PIMRC 2006, ‘Capacity evaluation of various multiuser MIMO schemes in downlink cellular environments’. Paper presented at The 17th Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, IEEE, PIMRC. Kirkebo, J. & Kiani, D, 2009 ‘Maximizing the capacity of wireless networks using multi-cell access schemes’. Paper presented at the Mobile Communications Department, Eurecom Institute, France. Lv, T & Wang, T, 2011 ‘Canceling Interferences for High Data Rate Time Reversal MIMO UWB System: A Precoding Approach’. EURASIP Journal on Wireless Communications and Networking 2011, Vol 2011. Mahomen, P., Petrova, M & Simic, L, 2009 ‘Wi-Fi, but not on Steroids: Performance Analysis of a Wi-Fi-like Network Operating in TVWS under Realistic Conditions’. Paper presented at the Institute for Networked Systems, RWTH Aachen University, Germany. Pahlavan, K & Zahedi, 2008 A. ‘Throughput of a wireless LAN access point in presence of Natural hidden terminals and capture effects’. Paper presented at the Center for Wireless Information Network Studies Worcester Polytechnic Institute, Worcester, MA 01 609 Understanding 802.11 Wi-Fi Access Points, 2012. Available from < http://www.connect802.com/access_points.htm> Read More