Advanced Block Acknowledgement and Cell Selection Techniques for Wireless Networks
|關鍵字:||區塊確認演算法;貪婪機制;起始序列編號;窗利用率;細胞切換機制;部分性觀察馬可夫決策過程;block acknowledgement;greedy scheme;starting sequence number;window utilization;cell selection;partially observable Markov decision process|
In this thesis, an advanced technique of the block acknowledgement will be illustrated first. The techniques of frame aggregation and block acknowledgement are utilized in the IEEE 802.11n standard for achieving high throughput performance. Conventional greedy scheme for block acknowledgement adopts the transmitter-defined starting sequence number (SSN) to construct the acknowledgement window for recognizing the correctness of data packets. However, there exists correctly received packets that lie outside of the acknowledgement window which will unavoidably be retransmitted. Therefore, a greedy fast-shift (GFS) block acknowledgement mechanism is proposed to provide the receiver-defined SSN, which can both implicitly acknowledge the correctly received packets before the SSN and explicitly identify the correctness information for the packets after the SSN. In order to evaluate the effectiveness of the GFS scheme, the analytical models for these two mechanisms are proposed based on the window utilization. It is observed from the simulation results that the proposed GFS method can provide better performance owing to its fast-shift behavior on acknowledgement window. In the second part of this thesis, an advanced cell selection mechanism is proposed with the consideration of available bandwidth of the base stations (BSs). The mobile station (MS) should execute the cell selection to choose a suitable BS for sustaining the quality of service (QoS) in the wireless networks. The conventional scheme of cell selection is based on the received signal strength (RSS) from the BS. However, the RSS-based scheme does not consider the available bandwidth allocation of the BSs, which is closely related to the capacity of the BSs. For those MSs that choose the serving BS (SBS) according to the maximum RSS criterion, there is no guarantee that maximum system capacity can be achieved. In order to acquire higher network capacity, the cell selection problem is formulated based on the partially observable Markov decision process (POMDP), which is designed to predict the unavailable capacity information from the non-serving BSs. Various utility functions are designed to consider different factors in the proposed POMDP-based cell selection (POCS) schemes, including the RSS, system capacity, handover time, and the MS's mobility. It is shown in the simulation results that the proposed POCS schemes can outperform existing RSS-based methods with higher system capacity.