Search

TGax aims at developing the IEEE 802.11ax protocol to provide enhanced throughput and power efficiency in dense WiFi deployment environments. The new specification shifts the focus from EDCA access to point coordination access, which aims at improving network utilization and power efficiency. Furthermore, it introduces, for the first time, the concept of RU allocation and orthogonal frequency division multiple access (OFDMA) techniques. The new specification addresses the spectral inefficiency in existing transmission by including schedule-based multi-user multiple-input, multiple- output (MU-MIMO) and multi-user OFDMA (MU-OFDMA) techniques to facilitate multi-user uplink transmission for high efficiency (HE) devices. However, complete adoption of the HE protocol is likely to take place over a period of time to offset the deployment cost. To ensure the coexistence of non-HE devices during the transition period, EDCA based single user transmission will remain the basic access technique and a focal point of spectral inefficiency for non-HE devices in the uplink direction. Motivated by our work on the performance of IEE 802.11ac, we propose an EDCA based uplink transmission technique for non-HE devices that allows multi-user transmission to improve spectral efficiency in coexistent WLAN networks.

The 802.11ac version of the popular IEEE 802.11 protocol aims to boost performance by increasing the channel bandwidth and allowing simultaneous transmission to multiple clients in the downlink direction. This dissertation presents an innovative approach of performance evaluation of downlink multi-user multiple input multiple output (DL-MUMIMO) technique for differentiated quality of service (QoS) based traffic categories in IEEE 802.11ac protocol. We propose a novel analytical model based on discrete Markov chain (DMC) and E-limited M/G/1 queuing model to evaluate the performance improvements of multi-user transmission opportunity (MU-TXOP) sharing in DL-MU-MIMO under non-saturated load and non-ideal channel condition. We also evaluate MU-TXOP sharing probabilities among different QoS nodes to assess the performance improvement of different traffic categories. In this dissertation, we propose an Access Point controlled MAC protocol (A-MAC) that enables simultaneous transmissions from multiple stations (STA) in uplink to eliminate under utilization of network resources in uplink transmission due to single user communication. The protocol uses enhanced distributed channel access (EDCA) technique to initiate multi-user transmission and orthogonal frequency division multiple access (OFDMA) method to transmit multiple Ready-To-Send (RTS) messages simultaneously. The proposed protocol also introduces explicit channel sounding technique by using dedicated OFDM subcarrier blocks for each user. We evaluate the performance of the proposed protocol using a noble analytical model and validate the performance metrics by extensive simulation in different traffic conditions. In this dissertation we also propose the increase of carrier sensing threshold (CSTH) of nodes during association with access point (AP) to mitigate collision probability due to hidden nodes during uplink transmission and validate our proposal through extensive simulation. Finally, we propose a noble analytical model to evaluate the performance of restricted access window (RAW) mechanism of IEEE 802.11ah as the MAC layer protocol for internet of things (IoT) network. We evaluate detailed performance metrics of non-QoS IoT network and investigate the feasibility of using RAW mechanism to support differentiated QoS based heterogeneous IoT network

The IEEE 802.11ac amendment enhances WLAN throughput by exploiting the spatial diversity of the antennas in a multi-user multi-input multi-output downlink transmission. Still, network resources remain under-utilized in uplink transmission due to single-user communication. In this paper, we propose an access point-controlled MAC protocol (A-MAC) that enables simultaneous transmissions from multiple STAs in uplink. The protocol uses the EDCA channel access technique to initiate multi-user transmission and the OFDMA method to transmit multiple RTSs simultaneously. It also introduces the explicit channel sounding technique by using dedicated OFDM subcarrier blocks for each user. Performance measurement shows that network throughput of the A-MAC is 150% higher than that of a single uplink transmission, thanks to the availability of concurrent multiple RTS transmissions in the uplink. The proposed protocol shortens the backoff time by up to 50% for all traffic categories due to concurrent multiple transmissions and thus enhances the battery life of the nodes. We observe that although the smaller backoff window of high-priority traffic category enhances the network throughput, higher intensity of high-priority traffic drives the network faster to saturation. Furthermore, better network stability and fairness among different traffic categories can be achieved when the dominant traffic has low priority.

The 802.11ac version of the popular IEEE 802.11 protocol aims to boost performance by increasing the channel bandwidth and allowing downlink multiuser multiple-input multiple-output (MIMO)—i.e., simultaneous transmission to multiple clients in the downlink direction. In this paper, we evaluate the performance improvements of multiuser Transmission Opportunity (MU-TXOP) sharing in this scenario under nonsaturated load and non-ideal channel condition, using a novel analytical model based on discrete Markov chain and E-limited M/G/1 queuing model. Our results show that MU-TXOP sharing achieves up to 105% gain in throughput compared to the case without MU-TXOP sharing. Moreover, the low-priority traffic benefits most from MU-TXOP sharing, with up to 150% throughput gain at low traffic intensity due to availability of additional transmission opportunities for secondary ACs.

IEEE 802.11ah protocol is specifically designed to provide network connectivity to a large number of energy efficient heterogeneous internet of things (IoT) devices. Restricted access window (RAW) mechanism of the protocol is an innovative feature which aims at reducing medium access contention by slotting the beacon interval and allowing limited number of nodes to contend in a specific slot. In this paper, we evaluate important medium access control (MAC) layer performance metrics of differentiated quality of service (QoS) heterogeneous IoT nodes in IEEE 802.11ah RAW mechanism. Our analysis evaluates the feasibility of coexistence of priority and non-priority traffic in IoT devices without degrading network performance.

IEEE 802.11ah protocol aims at providing network connectivity to a large number of end points while maintaining a satisfactory network performance. The RAW scheme of MAC access technique enables STAs to contend in different RAW slots and thereby reducing the chance of collision. In this paper we evaluate the performance metrics of the RAW scheme and observe that RAW scheme may be suitable for very small traffic condition. At higher traffic load, some of the nodes need to wait longer to transmit packets and network becomes unstable very quickly as the traffic for each node is increased beyond two packets per second in our test condition. The highest normalized network throughput achieved in this network is around 40%. Some important performance metrics like idle probability, backoff time and transmission probability are evaluated and studied in detail in this paper.

The IEEE 802.11ax draft of high efficiency (HE) amendment to the ubiquitous 802.11 standard aims at increasing spectral and power efficiency of the network. However, the performance degradation due to hidden terminal problem had always been a concern especially in uplink transmission. Although the trigger based uplink transmission in IEEE 802.11ax is successful to some extent, the hidden node problem still persists in coexisting network. In particular, the uplink throughput is drastically reduced with the increase of packet arrival rate in presence of hidden node. To combat the hidden terminal problem, we propose to increase the carrier sensing threshold (CSTH) of STAs during association with HE access point (AP). Our analysis shows that this approach allows collision probability in uplink transmission to be reduced significantly.

In this work we discuss architectural alternatives for the design of a proxy that interconnects IoT domains running CoAP with the rest of the Internet including micro datacenters and other domains building scalable hierarchical architectures. We assume that the CoAP domain is terminated by an IoT proxy with cache, and we investigate several design alternatives, assuming that the proxy autonomously maintains data freshness. Our analysis indicates that multicast and observe-based proxies perform better than the default POST/GET approach in terms of successful data transmission, round trip delay and energy consumption, with the multicast option having a slight advantage.

IEEE 802.11ac amendment enhances WLAN throughput by exploiting spatial diversity of antennas in MU-MIMO downlink transmission. However, in the existing framework ample network resources remain underutilized in uplink transmission due to single user communication. In this paper, we have proposed an efficient Access Point controlled MAC (A-MAC) protocol that enables concurrent multiple uplink transmissions for different priority traffic classes. Performance measurement shows that network throughput of A-MAC is 150\% higher than the throughput of single uplink transmission in existing IEEE 802.11ac protocol. We have achieve the highest payload throughput of 64\% when the number of nodes $N_{sum}=8$ and $\lambda=138$ packet/s at uniformly varying packet arrival rates for all STAs. As the number of nodes increases, the network throughput gradually decreases but the packet handling capacity of the network increases. The peak network capacity of 1200 packets/s is achieved with $N_{sum}=16$ and $\lambda=75$ packets/s for all STAs. Further increase of the number of nodes results in decreasing both network throughput and packet handling capacity of the network. We have also evaluated the performance metrics of our proposed A-MAC protocol under non-uniformly varying packet arrival rates for a fixed number of nodes $N_{sum}=24$. The analysis shows that when the network has higher intensity of low priority traffic, the aggregate packet handling capacity of the network is 1400 packets/s. However, as the intensity of high priority traffic increases, the aggregate packet handling capacity of the network decreases to 1000 packets/s. The results show that the more the higher priority packets arrive, the faster the network becomes unstable.

In this paper, we have explored the possibility of using IEEE 802.11ac as the MAC layer solution for VANET to meet the new requirements for various types of broadband services for Vehicle to Infrastructure (V2I) wireless communication. We have proposed an efficient RSU controlled MAC protocol that enables uplink MU-MIMO transmission for different priority traffic categories. Performance metrics show that the throughput enhancement of proposed MAC in MU-MIMO uplink transmission is three times the throughput of a single user transmission. We observe that due to smaller contention window size, the mean backoff time for the highest priority messages remain in the range of millisecond even at high packet arrival rate and node density. The saturation throughput analysis clearly shows two desired operating points: highest network throughput and highest network capacity. We achieve a maximum payload throughput of 64\% at an aggregate packet arrival rate of 1100 packets/sec whereas 54\% network throughput is achieved at a maximum load of 1200 packets/sec. We observe that the stable operating load remains between 1100-1200 packets/sec in our model.

IEEE 802.11ac standard provides an increase in performance by introducing the pioneering downlink multi-user MIMO technique which allows simultaneous transmission to multiple clients in the down-link direction [1]. In this paper we have evaluated MU-TXOP sharing probabilities in IEEE 802.11ac at different traffic intensities using analytical model based on discrete Markov chain and E-limited M/G/1 queuing model. Our results show that the low priority traffic group benefits most from MU-TXOP sharing and achieves 150% throughput gain at low traffic (12 Mbps).